WO2022057764A1 - 广告显示方法及电子设备 - Google Patents

广告显示方法及电子设备 Download PDF

Info

Publication number
WO2022057764A1
WO2022057764A1 PCT/CN2021/117991 CN2021117991W WO2022057764A1 WO 2022057764 A1 WO2022057764 A1 WO 2022057764A1 CN 2021117991 W CN2021117991 W CN 2021117991W WO 2022057764 A1 WO2022057764 A1 WO 2022057764A1
Authority
WO
WIPO (PCT)
Prior art keywords
electronic device
user
advertisement
knowledge
personal
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/117991
Other languages
English (en)
French (fr)
Inventor
刘璐
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to EP21868587.3A priority Critical patent/EP4202730A4/en
Priority to US18/027,040 priority patent/US12412195B2/en
Publication of WO2022057764A1 publication Critical patent/WO2022057764A1/zh
Anticipated expiration legal-status Critical
Priority to US19/302,163 priority patent/US20250371583A1/en
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0269—Targeted advertisements based on user profile or attribute
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0277—Online advertisement
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367—Ontology
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/02—Knowledge representation; Symbolic representation
    • G06N5/022—Knowledge engineering; Knowledge acquisition
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
    • G06Q30/0239—Online discounts or incentives
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0255—Targeted advertisements based on user history
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0269—Targeted advertisements based on user profile or attribute
    • G06Q30/0271—Personalized advertisement

Definitions

  • the present application relates to the technical field of data processing, and in particular, to an advertisement display method and an electronic device.
  • the delivery method of advertisements has formed a technology-based delivery model that takes the crowd as the delivery target and is product-oriented.
  • the server collects the business data of the user group.
  • the business data can be the type and duration of the advertisement watched by the user group, the operation record of closing the user group and ignoring the advertisement, etc.
  • the server performs group portrait for the user group according to the business data of the user group.
  • the result of the portrait can be which types of advertisements the user group watches the most, and which types of advertisements the user group is not interested in, etc.; the server filters multiple advertisements in the advertisement pool according to the results of the group portrait; the server will sort the advertisements sent to the user's electronic device.
  • the above advertisement delivery method utilizes the behavioral characteristics of the majority of user groups to push advertisements, but does not take into account the differences of individual users, such as individual users’ preferences and needs. The best advertising effect.
  • the present application provides an advertisement display method and an electronic device.
  • the combined advertising recommendation scheme of the terminal side and the server side is realized.
  • the advertising delivery effect of the advertising supplier is optimized, so that the advertising delivery of the advertising supplier is more accurate, and the economic benefit of the advertising supplier is improved.
  • the user's personal knowledge graph is constructed by using the personal data stored on the terminal side.
  • the user's personal knowledge graph can describe the user's behavioral characteristics in all aspects, and the user's personal knowledge graph is established on the terminal side, which protects the user. Security of private information.
  • the present application provides a method for displaying advertisements.
  • the method includes obtaining first personal data of a user by an electronic device, where the first personal data is the personal information of the user; the electronic device constructs personal knowledge according to the first personal data Graph; the personal knowledge graph includes the first personal data and the time when the first personal data was generated.
  • the electronic device obtains parameter information of the first advertisement content from the advertisement server, where the parameter information includes the type of the first advertisement content and the link address of the first advertisement content; obtained; the first advertisement content includes one or more advertisements.
  • the electronic device obtains the parameter information of the second advertisement content from the parameter information of the first advertisement content according to the personal knowledge graph; the electronic device obtains the second advertisement content according to the parameter information of the second advertisement content; the second advertisement content includes one or more an advertisement; the electronic device displays the second advertisement content on the display screen.
  • the electronic device may acquire the parameter information of the second advertisement content from the parameter information of the first advertisement content according to the personal knowledge graph in one or more of the following manners.
  • Mode 1 The electronic device retains the parameter information of all advertisements in the parameter information of the first advertisement content, and the electronic device just sorts the first advertisement content according to the predicted value of the user's preference according to the type of advertisements, and obtains the second advertisement content.
  • the parameter information of the advertisement content The parameter information of the advertisement content.
  • Manner 2 The electronic device selects the parameter information of a part of the advertisement from the parameter information of the first advertisement content to obtain the parameter information of the second advertisement content.
  • the electronic device sorts the content of the first advertisements according to the type of advertisements according to the predicted value of the user's favorite degree from high to low, and only retains the parameter information of the advertisement whose predicted value of the user's degree of likeability is higher than the first threshold, and obtains the second advertisement content.
  • the parameter information of the advertisement content is not limited to the content of the first advertisements according to the type of advertisements according to the predicted value of the user's favorite degree from high to low, and only retains the parameter information of the advertisement whose predicted value of the user's degree of likeability is higher than the first threshold, and obtains the second advertisement content.
  • the electronic device sends an advertisement recommendation request to the advertisement server, and the electronic device receives the parameter information of the first advertisement content returned by the advertisement server. After that, the electronic device further filters the parameter information of the first advertisement content to obtain the parameter information of the second advertisement content.
  • the electronic device uses the acquired personal data to construct a personal knowledge graph of the user, and trains a reordering model according to the personal knowledge graph.
  • the electronic device After the electronic device sends an advertisement recommendation request to the advertisement server, the electronic device receives the parameter information of the first advertisement content sent by the advertisement service server; after that, the electronic device further filters the parameter information of the first advertisement content according to the reordering model, and obtains For parameter information of the second advertisement content, the electronic device obtains the second advertisement content according to the parameter information of the second advertisement content, and recommends the second advertisement content to the user for viewing.
  • the method realizes the combined advertisement recommendation scheme of the terminal side and the server side.
  • the advertising delivery effect of the advertising supplier is optimized, so that the advertising delivery of the advertising supplier is more accurate, and the economic benefit of the advertising supplier is improved.
  • the user's personal knowledge graph is constructed by using the personal data stored on the terminal side.
  • the user's personal knowledge graph can describe the user's behavioral characteristics in all aspects, and the user's personal knowledge graph is established on the terminal side, which protects the user. Security of private information.
  • the electronic device constructs a personal knowledge graph according to the first personal data, which specifically includes: the electronic device obtains the second personal data from the first personal data;
  • the second personal data includes relational knowledge, event knowledge, and entity knowledge;
  • the electronic device stores relational knowledge, event knowledge, and entity knowledge in a predetermined structure;
  • the electronic device stores relational knowledge, event knowledge in a predetermined structure, and entity knowledge in a predetermined structure according to a predetermined structure
  • Build a user's personal knowledge graph the personal knowledge graph graphically displays the interconnected data structure between personal data; and the personal knowledge graph includes the first personal data and the time when the first personal data was generated, and the personal knowledge graph can represent personal data.
  • the relationship with time is convenient for subsequent electronic devices to update the personal knowledge graph according to time.
  • the first advertisement content is any one or more of the following: pictures, videos, text, and audio.
  • the first advertisement content may also include other content, which is not limited in this application.
  • the electronic obtains the first personal data of the user every fixed period.
  • the electronic device can acquire the new first personal data of the user at regular intervals, and add the new first personal data to the personal knowledge graph, thereby updating the personal data of the user in the personal knowledge graph.
  • the predetermined structure is a quintuple structure; the electronic device stores the relationship knowledge according to the predetermined structure, which specifically includes: the electronic device stores the relationship knowledge according to the first quintuple structure The structure is stored; the first quintuple structure is "first entity-relationship-second entity-first time point-first time interval"; the relationship represents the relationship between the first entity and the second entity, and the first time point is The time when the first entity establishes the relationship with the second entity, and the first time interval is the interval from the first time point to the current time point.
  • the electronic device stores the user's relational knowledge as a predetermined structure, which facilitates subsequent construction of a personal knowledge graph.
  • the first quintuple representing relational knowledge includes a first time point and a first time interval, and the electronic device can update the user's relational knowledge according to the first time point and the first time interval.
  • the predetermined structure is a quintuple structure; the electronic device stores the event knowledge according to the predetermined structure, which specifically includes: the electronic device stores the event knowledge according to the second quintuple structure structure for storage; the second quintuple structure is "event-argument-logical relationship-second time point-second time interval"; the argument is the occurrence action of the event, the logical relationship represents the relationship between the event and the argument, the first The second time point is the time when the event occurs, and the second time interval is the interval time from the second time point to the current time point.
  • the electronic device stores the user's event knowledge as a predetermined structure, which facilitates subsequent construction of a personal knowledge graph.
  • the second quintuple representing the event knowledge includes a second time point and a second time interval, and the electronic device can update the user's event knowledge according to the second time point and the second time interval.
  • the predetermined structure is a quintuple structure; the electronic device stores the entity knowledge according to the predetermined structure, which specifically includes: the electronic device stores the entity knowledge according to the third quintuple structure
  • the structure of the third quintuple is "third entity: the third time point - the first association weight - the fourth entity - the second association weight - the fifth entity"; the third time point is the occurrence of the third entity time, the first association weight is the association degree between the third entity and the fourth entity, and the second association weight is the association degree between the fourth entity and the fifth entity.
  • the electronic device stores the entity knowledge of the user as a predetermined structure, which facilitates subsequent construction of a personal knowledge graph.
  • the third quintuple representing the entity knowledge includes a third time point and a third time interval, and the electronic device can update the user's entity knowledge according to the third time point and the third time interval.
  • the electronic device deletes relational knowledge whose first time interval is greater than the first threshold in the personal knowledge graph; and/or, the electronic device deletes the second knowledge in the personal knowledge graph Event knowledge whose time interval is greater than the first threshold; and/or, the electronic device determines a third time interval from the third time point to the current time point according to the third time point; the electronic device deletes the third time interval in the personal knowledge graph greater than the third time interval.
  • a threshold of entity knowledge In this way, the electronic device can delete user knowledge whose time interval is greater than the first threshold in the personal knowledge graph according to time, and remove user knowledge from a long time ago.
  • the personal knowledge graph can better represent the behavior characteristics of the user in the recent period.
  • the method further includes: the electronic device obtains the user's historical behavior and the electronic device display The electronic device takes the historical advertising information and personal knowledge graph as the input of the reordering model, and the reordering model outputs the first result; the electronic device compares the first result with the user's historical behavior, and modifies the parameters of the reordering model, Until the first result output by the reordering model is within the preset range, the first model is obtained; the electronic device obtains the parameter information of the second advertisement content from the parameter information of the first advertisement content according to the personal knowledge graph, and specifically includes: the electronic device The parameter information of the second advertisement content is acquired from the parameter information of the first advertisement content according to the first model. In this way, the electronic device trains the reordering model according to the personal knowledge graph to obtain the first model.
  • the electronic device may acquire the parameter information of the second advertisement content from the parameter information
  • the electronic device obtains the parameter information of the second advertisement content from the parameter information of the first advertisement content according to the first model, which specifically includes: the electronic device obtains the parameter information of the second advertisement content according to the first model.
  • a model sorts the types of the first advertisement content from high to low according to the predicted value of the user's degree of preference, and obtains parameter information of the second advertisement content; The likeness predicted value is sorted from high to low, and the types of advertisements whose likeability predicted value of the user is higher than the first threshold value are acquired, and the parameter information of the second advertisement content is obtained.
  • the electronic device obtains the parameter information of the second advertisement content from the parameter information of the first advertisement content according to the user's preference prediction value, so that the advertisement displayed by the electronic device is more in line with the user's preference. In this way, the advertisement recommendation effect can be improved.
  • the method further includes: the electronic device converts the first personal data into text information; the electronic device performs sentence segmentation, word segmentation and part-of-speech tagging on the text information; the electronic device obtains the second personal data from the first personal data, which specifically includes: the electronic device obtains words belonging to a preset part of speech in the text information.
  • the electronic device removes data that cannot characterize the user's behavior in the first personal data.
  • the electronic device removes useless data, and the obtained second personal data can better describe the user's behavioral characteristics, so that the constructed personal knowledge graph can more accurately represent the user's behavioral characteristics.
  • the method further includes: the electronic device acquires words that appear once in the text information; If there are two or more identical words in the text information, the electronic device obtains one of the two or more identical words in the text information, and obtains the second personal data. In this way, the electronic device removes duplicate data and reduces data redundancy.
  • the user's personal information includes one or more of the following: gender, age, personality, hobbies, interpersonal relationships, income, address book information, call records, Text messages, memo messages, residential addresses, weather conditions of residential addresses.
  • the electronic device displays the second advertisement content in the advertisement display area of the display screen, which specifically includes: the electronic device predicts a value according to the user's preference in the second advertisement content Plays one or more advertisements in the second advertisement content from high to low; or, the electronic device plays the advertisement with the highest predicted value of the user's favorite degree in the second advertisement content; Playing one or more advertisements in the advertisements from high to low with the predicted favorite degree, and blocking one or more advertisements in the second advertisement content that were played by the electronic device in the first time period.
  • the electronic device plays one or more advertisements in the second advertisement content from high to low according to the predicted value of the user's degree of preference, or plays the advertisement with the highest predicted value of the user's degree of preference, which is more in line with the user's preference. The higher the probability.
  • the electronic device blocks one or more advertisements played by the electronic device in the first time period, so as to avoid recommending the same advertisement repeatedly in a short period of time, which affects the user experience.
  • the method further includes: the electronic device obtains viewing data of the second advertisement content by the user
  • the viewing data includes the advertisement type of one or more advertisements in the second advertisement content that the user watched and the advertisement type of which the user closed one or more advertisements in the second advertisement content; the electronic device updates the first model according to the viewing data.
  • the electronic device updates the first model according to the advertisement viewing data of the user, and the first model recommends the advertisement of the type most viewed by the user to the user when recommending advertisements to the user next time, which is more in line with the user's needs.
  • the present application provides an electronic device, the electronic device includes one or more processors, one or more memories, and a display screen; the one or more memories, the display screen are coupled with the one or more processors, and one One or more memories are used to store computer program code, the computer program code includes computer instructions, and one or more processors invoke the computer instructions to cause the electronic device to perform the above-mentioned first aspect and any one of the implementation manners of the above-mentioned first aspect.
  • the present application provides a computer storage medium, where the computer readable storage medium stores a computer program.
  • the processor executes the first aspect and any one of the first aspect in combination with the processor.
  • An advertisement display method provided by an implementation manner.
  • an embodiment of the present application provides a computer program product, where a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the first aspect above and in combination with the above-mentioned first aspect An advertisement display method provided by any implementation manner of .
  • the electronic device uses the acquired personal data to construct a personal knowledge graph of the user, and trains a reordering model according to the personal knowledge graph.
  • the electronic device After the electronic device sends an advertisement recommendation request to the advertisement server, the electronic device receives the parameter information of the first advertisement content sent by the advertisement service server; after that, the electronic device further filters the parameter information of the first advertisement content according to the reordering model, and obtains the first advertisement content.
  • the parameter information of the advertisement content The electronic device obtains the second advertisement content according to the parameter information of the second advertisement content, and recommends the second advertisement content to the user for viewing.
  • the method realizes the combined advertisement recommendation scheme of the terminal side and the server side.
  • the advertising delivery effect of the advertising supplier is optimized, so that the advertising delivery of the advertising supplier is more accurate, and the economic benefit of the advertising supplier is improved.
  • the user's personal knowledge graph is constructed by using the personal data stored on the terminal side.
  • the user's personal knowledge graph can describe the user's behavioral characteristics in all aspects, and the user's personal knowledge graph is established on the terminal side, which protects the user. Security of private information.
  • FIG. 1 is a schematic diagram of an advertisement recommendation system provided by an embodiment of the present application
  • FIG. 2 is a schematic structural diagram of an electronic device 100 according to an embodiment of the present application.
  • FIG. 3 is a block diagram of a software structure of an electronic device 100 according to an embodiment of the present application.
  • FIG. 4 is a schematic diagram of a hardware structure of an advertisement server 200 according to an embodiment of the present application.
  • FIG. 5 is a schematic diagram of the architecture of another advertisement recommendation system provided by an embodiment of the present application.
  • 6A-6E are a set of application program interface diagrams provided by the embodiments of the present application.
  • FIG. 7 is a schematic flowchart of an advertisement display method provided by an embodiment of the present application.
  • FIG. 8 is a schematic diagram representing a constructed personal knowledge graph in the form of a graph according to an embodiment of the present application.
  • 9A-9C are a set of UI diagrams provided by the embodiments of the present application.
  • FIG. 10 is a schematic flowchart of another advertisement display method provided by an embodiment of the present application.
  • FIGS. 11A-11C are another set of UI diagrams provided by this embodiment of the present application.
  • FIG. 12 is a schematic diagram of another system architecture provided by an embodiment of the present application.
  • FIG. 12A is a UI diagram provided by an embodiment of the present application.
  • first and second are only used for descriptive purposes, and should not be construed as implying or implying relative importance or implying the number of indicated technical features. Therefore, the features defined as “first” and “second” may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present application, unless otherwise specified, the “multiple” The meaning is two or more.
  • UI user interface
  • the term "user interface (UI)" in the description, claims and drawings of this application is a medium interface for interaction and information exchange between an application program or an operating system and a user, and it realizes the internal form of information Conversion to and from user-acceptable forms.
  • the user interface of the application is the source code written in a specific computer language such as java and extensible markup language (XML).
  • the interface source code is parsed and rendered on the terminal device, and finally presented as content that the user can recognize.
  • Controls also known as widgets, are the basic elements of the user interface. Typical controls include toolbars, menu bars, text boxes, buttons, and scroll bars. (scrollbar), pictures and text.
  • the attributes and content of controls in the interface are defined by tags or nodes.
  • XML specifies the controls contained in the interface through nodes such as ⁇ Textview>, ⁇ ImgView>, and ⁇ VideoView>.
  • a node corresponds to a control or property in the interface, and the node is rendered as user-visible content after parsing and rendering.
  • applications such as hybrid applications, often contain web pages in the interface.
  • a web page, also known as a page can be understood as a special control embedded in an application program interface.
  • a web page is source code written in a specific computer language, such as hypertext markup language (HTML), cascading styles Tables (cascading style sheets, CSS), java scripts (JavaScript, JS), etc.
  • the source code of the web page can be loaded and displayed as user-identifiable content by a browser or a web page display component similar in function to a browser.
  • the specific content contained in a web page is also defined by tags or nodes in the source code of the web page. For example, HTML defines the elements and attributes of web pages through ⁇ p>, ⁇ img>, ⁇ video>, and ⁇ canvas>.
  • GUI graphical user interface
  • GUI refers to a user interface related to computer operations that is displayed graphically. It can be an icon, window, control and other interface elements displayed on the display screen of the electronic device, wherein the control can include icons, buttons, menus, tabs, text boxes, dialog boxes, status bars, navigation bars, Widgets, etc. visual interface elements.
  • Unstructured personal data is data that cannot be represented by a two-dimensional logical table.
  • Unstructured personal data may be data such as documents, pictures, videos, texts, and the like.
  • the unstructured personal data may be data generated during the operation of the camera application.
  • Pictures and videos shot by the camera app are stored in the file system, so the pictures and videos shot by the camera app are unstructured personal data.
  • Data services are used to store structured personal data generated during the operation of various applications in electronic devices.
  • Structured personal data is data that can be represented by a unified structure.
  • the structured personal data may be data generated during the operation of the above-mentioned address book application.
  • the user contact name and the user contact phone number stored in the address book application are stored in a one-to-one correspondence between the user contact name and the user contact phone number in the data service.
  • User contact names and user contact phone numbers are structured personal data.
  • Personal data includes data concerning the privacy of individuals.
  • the personal data may be data related to personal privacy generated during the process of running various application programs on the electronic device, and the data generated during the process of each application program is saved in the file system and/or data service.
  • personal data can also be data involving personal privacy that the electronic device obtains directly from various applications after obtaining the authorization of the user. For example, communication applications, SMS applications, address book applications, memo applications, weather applications, shopping applications, etc.
  • the data generated by each application program in the electronic device is stored in the data service and/or the file system, and the electronic device can obtain the user's personal data from the data service and/or the file system.
  • the application in the electronic device can obtain the user's authorization, and after obtaining the user's authorization, the electronic device can obtain the user's personal data from each application.
  • the personal data obtained by the electronic device directly from the application program can also be divided into structured personal data and unstructured personal data.
  • Group data includes data of multiple users in the user group that do not involve user privacy, such as business data generated when users watch advertisements. For example, it may include one or more of the following: advertisements that the user often clicks, advertisements that the user has never clicked, the time that the user watches the advertisement, the advertisement that the user closes, and so on.
  • the personal knowledge graph is constructed based on the user's personal data, and is a data structure that graphically displays the interconnectedness of personal data.
  • the group knowledge graph is constructed according to the group data of the user group, that is, the knowledge graph is a data structure representing the interrelationship between the group data of the user group.
  • the group knowledge graph cannot represent the behavioral characteristics of individual users.
  • the electronic device may construct the personal knowledge graph of each user according to the personal data of the user.
  • the specific construction process of the personal knowledge graph reference may be made to the detailed description of the subsequent method embodiments, which will not be repeated for the time being.
  • Group portraits are labels for user groups generated using group data.
  • the tags of the user group may include, but are not limited to, the types of advertisements that the user group likes to browse, the types of advertisements that the user group ignores the most, the types of advertisements that the user group closes the most ads, and the types of advertisements that the user group reports the most ads.
  • a user population is a set of all users regardless of gender, age, or region.
  • the user group may be classified according to gender, for example, the user group may be divided into a female user group and a male user group. Alternatively, the user group may also be divided into user groups of various age groups, and the like.
  • the group portrait can be divided into a female user group portrait and a male user group portrait.
  • the group data can be divided into data of male user groups and data of female user groups.
  • the male user group is profiled according to the data of the male user group, that is, the male user group is labeled.
  • the male user group is most interested in car advertisements, but not interested in beauty advertisements.
  • the female user group is profiled according to the data of the female user group, that is, the female user group is labeled.
  • the female user group is interested in clothing advertisements and beauty advertisements, but not interested in sports advertisements, etc.
  • the group portrait can also be used for the group portrait of the user group according to the user groups of each age group.
  • the group data can be classified according to age groups. For example, the user group data with the user age between 0-20 is divided into one category, the user group data with the user age between 21-35 is divided into one category, and the user age is between 36-50.
  • the user group data is divided into one category, and the user group data between the ages of 51 and 70 is divided into one category.
  • the user groups of each age group are profiled according to the user group data of each age group, that is, the user groups of each age group are labeled.
  • the user group between the ages of 0-20 is most interested in toy advertisements
  • the user group between the ages of 21-35 is most interested in the advertisement of electronic products
  • the user group between the ages of 36-50 is most interested in advertisements of electronic products.
  • User groups are most interested in skin care and hair care advertisements
  • user groups between the ages of 51-70 are most interested in health care advertisements.
  • Advertising is a means of disseminating information to the masses. Advertising can be divided into public service advertising and profit advertising. Public service advertisements are non-profit advertising activities that provide free services to the society. Profitable advertising can include the promotion of various applications, products or some brands. For example, the profitable advertisements may be beauty makeup, food, category advertisements, music audition recommendation advertisements, video recommendation advertisements, novel recommendation advertisements, movie recommendation advertisements, application download recommendation advertisements, and so on.
  • the ad serving process can include the following steps:
  • the advertisement server receives an advertisement recommendation request sent by the application server, and in response to the advertisement recommendation request, the advertisement server sends an advertisement acquisition request to a manufacturer server (eg, a mobile phone manufacturer server).
  • a manufacturer server eg, a mobile phone manufacturer server
  • the role of the advertisement server is to obtain advertisements of various manufacturers and screen the advertisements, and then the advertisement server sends the filtered advertisements to each application program for display.
  • any vendor server can send the ad to be sent to the ad server.
  • the advertisement server receives advertisements sent by any manufacturer's server, and the advertisement server filters multiple advertisements (for example, according to the advertisement prices from high to low) to obtain an advertisement set.
  • advertisements sent by any manufacturer server to the advertisement server may be duplicated, and the advertisement server may also de-duplicate the acquired advertisements to avoid duplicate advertisements.
  • the advertisement server further filters the advertisements in the advertisement set according to the group data of the user group to obtain the advertisement list, the advertisement server sends the advertisement list to the application program, and the application program displays the advertisements in the advertisement list.
  • Roughly arranged advertisements are a collection of one or more advertisements. Roughly arranged advertisements are obtained by the advertisement server filtering a large number of advertisements according to the group data according to the interest degree of the user groups in the advertisements.
  • Placement of advertisements is a collection of one or more advertisements, which are screened by electronic devices from the rough-arranged advertisements according to the user's personal data and the probability that individuals may click on the advertisements.
  • the number of advertisements placed may be the same as the number of rough-arranged advertisements.
  • the number of advertisements placed may also be smaller than the number of coarse-ranked advertisements, because the electronic device can filter out some advertisements in the coarse-ranked advertisements according to the user's personal data.
  • FIG. 1 is a schematic diagram of an advertisement recommendation system.
  • the system 10 may include the electronic device 100 shown in FIG. 1 , an advertisement server 200 and an application server 300 .
  • the electronic device 100 may be used to detect a user operation, and in response to the user operation, the electronic device 100 sends a user request to the application server 300 .
  • the user operation may be, for example: the electronic device 100 detects that the user has opened the application, or the electronic device 100 detects that the user next video is the current browsing page of the application (for example, the electronic device 100 detects that the user swipes down with a single finger) The next video is currently browsing interface), the electronic device 100 sends a user request to the application server 300 .
  • the application server 300 may be configured to receive and respond to user requests, and the application server 300 sends an advertisement recommendation request to the advertisement server 200 .
  • the advertisement server 200 may be configured to screen the advertisements in the advertisement set according to the group portrait of the user, so as to obtain coarsely arranged advertisements.
  • the advertisement server 200 can also be used to send the rough-form advertisement to the electronic device 100 .
  • the electronic device 100 can also be used to receive the rough-arranged advertisements sent by the advertisement server 200, and further filter the rough-arranged advertisements to obtain the advertisements to be placed. After that, the electronic device 100 recommends the advertisements to be placed to the user for viewing.
  • the architecture of the advertisement recommendation system in FIG. 1 is only an exemplary implementation in the embodiment of the present application, and the architecture of the advertisement recommendation system in the embodiment of the present application includes but is not limited to the above advertisement recommendation system architecture.
  • Method 1 The application server will collect the business data of multiple users watching advertisements, and upload the business data of the user group to the advertisement server.
  • the advertisement server screens multiple advertisements according to the business data of the user group, and obtains a list of advertisements that the user group is interested in.
  • the ad server sends advertisements that are of interest to the user community to the user.
  • This advertising delivery method utilizes the business data of user groups to advertise. This advertising delivery method does not take into account the differences of individual users, and there is a problem of homogenization of recommendations.
  • Method 2 The application server will collect the search information and browsing information of individual users, and the application server will extract the keywords in the search information and browsing information to recommend content to individual users.
  • an individual user searches a shopping application for an item he wants to buy (eg, earphones), and the shopping application recommends a variety of earphone item information to the individual user the next time the individual user uses the shopping application.
  • This recommendation method will collect the user's search information and browsing information, which will lead to the leakage of the user's personal privacy for the user.
  • the electronic device can receive a user operation, and in response to the user operation, the electronic device sends a user request to the application server, the application server sends an advertisement recommendation request to the advertisement server, and the electronic device receives the rough layout advertisement (first advertisement content). After that, the electronic device further screens the rough-arranged advertisements to obtain the advertisements (second advertisement content).
  • the number of rough layout advertisements and advertisements can be one or more.
  • the electronic device may also directly receive the rough-formed advertisement (first advertisement content) returned by the advertisement server. After that, the electronic device further screens the rough-arranged advertisements to obtain the advertisements (second advertisement content). The electronic device does not have to send a user request to the application server. Please do not limit yourself here.
  • the electronic device uses the acquired personal data to construct a personal knowledge graph of the user, and trains a reordering model according to the personal knowledge graph.
  • the application server sends an advertisement recommendation request to the advertisement server
  • the electronic device receives the coarse-arranged advertisements sent by the advertising service server; after that, the electronic device further filters the coarse-arranged advertisements according to the reordering model to obtain the advertisements to be placed, and the electronic devices will place the advertisements Ads are recommended for users to watch.
  • the electronic device receives parameter information from the advertisement server, which may be the first advertisement content.
  • the parameter information may be information such as the type, link address, and size of the first advertisement content.
  • the electronic device obtains the parameter information of the second advertisement content from the parameter information of the first advertisement content according to the personal knowledge graph.
  • the electronic device acquires the second advertisement content from the advertisement server according to the parameter information of the second advertisement content. This application is not limited here.
  • the method realizes the combined advertisement recommendation scheme of the terminal side and the server side.
  • the advertising delivery effect of the advertising supplier is optimized, so that the advertising delivery of the advertising supplier is more accurate, and the economic benefit of the advertising supplier is improved.
  • the user's personal knowledge graph is constructed by using the personal data stored on the terminal side.
  • the user's personal knowledge graph can describe the user's behavioral characteristics in all aspects, and the user's personal knowledge graph is established on the terminal side, which protects the user. Security of private information.
  • FIG. 2 shows a schematic structural diagram of the electronic device 100 .
  • Device types of electronic device 100 may include cell phones, televisions, tablets, speakers, watches, desktop computers, laptop computers, handheld computers, notebook computers, ultra-mobile personal computers (UMPCs), netbooks, and personal digital Assistant (personal digital assistant, PDA), augmented reality (augmented reality, AR)/virtual reality (virtual reality, VR) equipment, etc.
  • PDA personal digital assistant
  • augmented reality augmented reality, AR
  • virtual reality virtual reality, VR
  • the electronic device 100 shown in FIG. 2 is only an example, and the electronic device 100 may have more or fewer components than those shown in FIG. 2, two or more components may be combined, or Different component configurations are possible.
  • the various components shown in the figures may be implemented in hardware, software, or a combination of hardware and software, including one or more signal processing and/or application specific integrated circuits.
  • the electronic device 100 may include: a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2.
  • Mobile communication module 150 wireless communication module 160, audio module 170, speaker 170A, receiver 170B, microphone 170C, headphone jack 170D, sensor module 180, buttons 190, motor 191, indicator 192, camera 193, display screen 194, And a subscriber identification module (subscriber identification module, SIM) card interface 195 and so on.
  • SIM subscriber identification module
  • the sensor module 180 may include a pressure sensor 180A, a gyroscope sensor 180B, an air pressure sensor 180C, a magnetic sensor 180D, an acceleration sensor 180E, a distance sensor 180F, a proximity light sensor 180G, a fingerprint sensor 180H, a temperature sensor 180J, a touch sensor 180K, and ambient light. Sensor 180L, bone conduction sensor 180M, etc.
  • the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on the electronic device 100 .
  • the electronic device 100 may include more or less components than shown, or combine some components, or separate some components, or arrange different components.
  • the illustrated components may be implemented in hardware, software, or a combination of software and hardware.
  • the processor 110 may include one or more processing units, for example, the processor 110 may include an application processor (application processor, AP), a modem processor, a graphics processor (graphics processing unit, GPU), an image signal processor (image signal processor, ISP), controller, memory, video codec, digital signal processor (digital signal processor, DSP), baseband processor, or neural network processor (neural-network processing unit, NPU), etc. Wherein, different processing units may be independent devices, or may be integrated in one or more processors.
  • application processor application processor, AP
  • modem processor graphics processor
  • ISP image signal processor
  • controller memory
  • video codec digital signal processor
  • DSP digital signal processor
  • baseband processor baseband processor
  • neural network processor neural-network processing unit
  • the controller may be the nerve center and command center of the electronic device 100 .
  • the controller can generate an operation control signal according to the instruction operation code and timing signal, and complete the control of fetching and executing instructions.
  • a memory may also be provided in the processor 110 for storing instructions and data.
  • the memory in processor 110 is cache memory. This memory may hold instructions or data that have just been used or recycled by the processor 110 . If the processor 110 needs to use the instruction or data again, it can be called directly from memory. Repeated accesses are avoided and the latency of the processor 110 is reduced, thereby increasing the efficiency of the system.
  • the processor 110 may include one or more interfaces.
  • the interface may include an integrated circuit (inter-integrated circuit, I2C) interface, an integrated circuit built-in audio (inter-integrated circuit sound, I2S) interface, a pulse code modulation (pulse code modulation, PCM) interface, a universal asynchronous transceiver (universal asynchronous transmitter) receiver/transmitter, UART) interface, mobile industry processor interface (MIPI), general-purpose input/output (GPIO) interface, subscriber identity module (SIM) interface, or Universal serial bus (universal serial bus, USB) interface, etc.
  • I2C integrated circuit
  • I2S integrated circuit built-in audio
  • PCM pulse code modulation
  • PCM pulse code modulation
  • UART universal asynchronous transceiver
  • MIPI mobile industry processor interface
  • GPIO general-purpose input/output
  • SIM subscriber identity module
  • USB Universal serial bus
  • the charging management module 140 is used to receive charging input from the charger.
  • the charger may be a wireless charger or a wired charger.
  • the charging management module 140 may receive charging input from the wired charger through the USB interface 130 .
  • the charging management module 140 may receive wireless charging input through a wireless charging coil of the electronic device 100 . While the charging management module 140 charges the battery 142 , it can also supply power to the electronic device through the power management module 141 .
  • the power management module 141 is used for connecting the battery 142 , the charging management module 140 and the processor 110 .
  • the power management module 141 receives input from the battery 142 and/or the charging management module 140 and supplies power to the processor 110 , the internal memory 121 , the external memory, the display screen 194 , the camera 193 , and the wireless communication module 160 .
  • the power management module 141 can also be used to monitor parameters such as battery capacity, battery cycle times, battery health status (leakage, impedance).
  • the power management module 141 may also be provided in the processor 110 .
  • the power management module 141 and the charging management module 140 may also be provided in the same device.
  • the wireless communication function of the electronic device 100 may be implemented by the antenna 1, the antenna 2, the mobile communication module 150, the wireless communication module 160, the modulation and demodulation processor, the baseband processor, and the like.
  • Antenna 1 and Antenna 2 are used to transmit and receive electromagnetic wave signals.
  • Each antenna in electronic device 100 may be used to cover a single or multiple communication frequency bands. Different antennas can also be reused to improve antenna utilization.
  • the mobile communication module 150 may provide wireless communication solutions including 2G/3G/4G/5G etc. applied on the electronic device 100 .
  • the mobile communication module 150 may include at least one filter, switch, power amplifier, low noise amplifier (LNA) and the like.
  • the mobile communication module 150 can receive electromagnetic waves from the antenna 1, filter and amplify the received electromagnetic waves, and transmit them to the modulation and demodulation processor for demodulation.
  • the mobile communication module 150 can also amplify the signal modulated by the modulation and demodulation processor, and then turn it into an electromagnetic wave for radiation through the antenna 1 .
  • the modem processor may include a modulator and a demodulator.
  • the modulator is used to modulate the low frequency baseband signal to be sent into a medium and high frequency signal.
  • the demodulator is used to demodulate the received electromagnetic wave signal into a low frequency baseband signal.
  • the wireless communication module 160 can provide applications on the electronic device 100 including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), bluetooth (BT), global navigation satellites Wireless communication solutions such as global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), and infrared technology (IR).
  • WLAN wireless local area networks
  • BT Bluetooth
  • GNSS global navigation satellite system
  • FM frequency modulation
  • NFC near field communication
  • IR infrared technology
  • the wireless communication module 160 may be one or more devices integrating at least one communication processing module.
  • the wireless communication module 160 receives electromagnetic waves via the antenna 2 , frequency modulates and filters the electromagnetic wave signals, and sends the processed signals to the processor 110 .
  • the wireless communication module 160 can also receive the signal to be sent from the processor 110 , perform frequency modulation on it, amplify it, and convert it into electromagnetic waves for radiation through the antenna 2 .
  • the electronic device 100 implements a display function through a GPU, a display screen 194, an application processor, and the like.
  • Display screen 194 is used to display images, videos, and the like.
  • Display screen 194 includes a display panel.
  • the electronic device 100 may implement a shooting function through an ISP, a camera 193, a video codec, a GPU, a display screen 194, an application processor, and the like.
  • the wireless communication solution provided by the mobile communication module 150 may enable the electronic device to communicate with a device in the network (such as an advertisement server), and the WLAN wireless communication solution provided by the wireless communication module 160 may also enable the electronic device Devices can communicate with devices in the network, such as ad servers.
  • the electronic device 100 may send an advertisement recommendation request through the wireless communication module 160 to establish a communication connection with the advertisement server, and the electronic device 100 may also receive the rough advertisement sent by the advertisement server through the wireless communication module 160 , and the electronic device 100 may further
  • the coarsely arranged advertisements can be screened by the processor 110 to obtain the advertisements to be placed, and the electronic device 100 can also be used to display the placed advertisements through the display screen 194 for the user to watch.
  • the ISP is used to process the data fed back by the camera 193 .
  • the shutter is opened, the light is transmitted to the camera photosensitive element through the lens, the light signal is converted into an electrical signal, and the camera photosensitive element transmits the electrical signal to the ISP for processing, converting it into an image visible to the naked eye.
  • Camera 193 is used to capture still images or video.
  • the object is projected through the lens to generate an optical image onto the photosensitive element.
  • a digital signal processor is used to process digital signals, in addition to processing digital image signals, it can also process other digital signals.
  • the NPU is a neural-network (NN) computing processor.
  • NN neural-network
  • the external memory interface 120 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the electronic device 100 .
  • an external memory card such as a Micro SD card
  • Internal memory 121 may be used to store computer executable program code, which includes instructions.
  • the processor 110 executes various functional applications and data processing of the electronic device 100 by executing the instructions stored in the internal memory 121 .
  • the electronic device 100 may implement audio functions through an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, an earphone interface 170D, an application processor, and the like. Such as music playback, recording, etc.
  • the audio module 170 is used for converting digital audio information into analog audio signal output, and also for converting analog audio input into digital audio signal. Audio module 170 may also be used to encode and decode audio signals.
  • Speaker 170A also referred to as a "speaker" is used to convert audio electrical signals into sound signals.
  • the electronic device 100 can listen to music through the speaker 170A, or listen to a hands-free call.
  • the receiver 170B also referred to as "earpiece" is used to convert audio electrical signals into sound signals.
  • the voice can be answered by placing the receiver 170B close to the human ear.
  • the microphone 170C also called “microphone” or “microphone” is used to convert sound signals into electrical signals.
  • the user can make a sound by approaching the microphone 170C through a human mouth, and input the sound signal into the microphone 170C.
  • the electronic device 100 may be provided with at least one microphone 170C. In other embodiments, the electronic device 100 may be provided with two microphones 170C, which can implement a noise reduction function in addition to collecting sound signals. In other embodiments, the electronic device 100 may further be provided with three, four or more microphones 170C to collect sound signals, reduce noise, identify sound sources, and implement directional recording functions.
  • the electronic device 100 collects the sound signal through the microphone 170C, and transmits the sound signal to the application program in the electronic device 100 .
  • the earphone jack 170D is used to connect wired earphones.
  • the earphone interface 170D can be the USB interface 130, or can be a 3.5mm open mobile terminal platform (OMTP) standard interface, a cellular telecommunications industry association of the USA (CTIA) standard interface.
  • OMTP open mobile terminal platform
  • CTIA cellular telecommunications industry association of the USA
  • the pressure sensor 180A is used to sense pressure signals, and can convert the pressure signals into electrical signals.
  • the pressure sensor 180A may be provided on the display screen 194 .
  • the capacitive pressure sensor may be comprised of at least two parallel plates of conductive material. When a force is applied to the pressure sensor 180A, the capacitance between the electrodes changes.
  • the electronic device 100 determines the intensity of the pressure according to the change in capacitance. When a touch operation acts on the display screen 194, the electronic device 100 detects the intensity of the touch operation according to the pressure sensor 180A.
  • the electronic device 100 may also calculate the touched position according to the detection signal of the pressure sensor 180A.
  • touch operations acting on the same touch position but with different touch operation intensities may correspond to different operation instructions. For example, when a touch operation whose intensity is less than the first pressure threshold acts on the short message application icon, the instruction for viewing the short message is executed. When a touch operation with a touch operation intensity greater than or equal to the first pressure threshold acts on the short message application icon, the instruction to create a new short message is executed.
  • the gyro sensor 180B may be used to determine the motion attitude of the electronic device 100 .
  • the angular velocity of electronic device 100 about three axes ie, x, y, and z axes
  • the gyro sensor 180B can be used for image stabilization.
  • the gyro sensor 180B detects the shaking angle of the electronic device 100, calculates the distance that the lens module needs to compensate according to the angle, and allows the lens to offset the shaking of the electronic device 100 through reverse motion to achieve anti-shake.
  • the gyro sensor 180B can also be used for navigation and somatosensory game scenarios.
  • the air pressure sensor 180C is used to measure air pressure.
  • the electronic device 100 calculates the altitude through the air pressure value measured by the air pressure sensor 180C to assist in positioning and navigation.
  • the magnetic sensor 180D includes a Hall sensor.
  • the electronic device 100 can detect the opening and closing of the flip holster using the magnetic sensor 180D.
  • the electronic device 100 can detect the opening and closing of the flip according to the magnetic sensor 180D. Further, according to the detected opening and closing state of the leather case or the opening and closing state of the flip cover, characteristics such as automatic unlocking of the flip cover are set.
  • the acceleration sensor 180E can detect the magnitude of the acceleration of the electronic device 100 in various directions (generally three axes).
  • the magnitude and direction of gravity can be detected when the electronic device 100 is stationary. It can also be used to identify the posture of electronic devices, and can be used in horizontal and vertical screen switching, pedometers and other applications.
  • the electronic device 100 can measure the distance through infrared or laser. In some embodiments, when shooting a scene, the electronic device 100 can use the distance sensor 180F to measure the distance to achieve fast focusing.
  • Proximity light sensor 180G may include, for example, light emitting diodes (LEDs) and light detectors, such as photodiodes.
  • the light emitting diodes may be infrared light emitting diodes.
  • the electronic device 100 emits infrared light to the outside through the light emitting diode.
  • Electronic device 100 uses photodiodes to detect infrared reflected light from nearby objects. When sufficient reflected light is detected, it can be determined that there is an object near the electronic device 100 . When insufficient reflected light is detected, the electronic device 100 may determine that there is no object near the electronic device 100 .
  • the electronic device 100 can use the proximity light sensor 180G to detect that the user holds the electronic device 100 close to the ear to talk, so as to automatically turn off the screen to save power.
  • Proximity light sensor 180G can also be used in holster mode, pocket mode automatically unlocks and locks the screen.
  • the ambient light sensor 180L is used to sense ambient light brightness.
  • the electronic device 100 can adaptively adjust the brightness of the display screen 194 according to the perceived ambient light brightness.
  • the ambient light sensor 180L can also be used to automatically adjust the white balance when taking pictures.
  • the ambient light sensor 180L can also cooperate with the proximity light sensor 180G to detect whether the electronic device 100 is in a pocket, so as to prevent accidental touch.
  • the fingerprint sensor 180H is used to collect fingerprints.
  • the electronic device 100 can use the collected fingerprint characteristics to realize fingerprint unlocking, accessing application locks, taking pictures with fingerprints, answering incoming calls with fingerprints, and the like.
  • the temperature sensor 180J is used to detect the temperature.
  • the electronic device 100 uses the temperature detected by the temperature sensor 180J to execute a temperature processing strategy. For example, when the temperature reported by the temperature sensor 180J exceeds a threshold value, the electronic device 100 reduces the performance of the processor located near the temperature sensor 180J in order to reduce power consumption and implement thermal protection.
  • the electronic device 100 when the temperature is lower than another threshold, the electronic device 100 heats the battery 142 to avoid abnormal shutdown of the electronic device 100 caused by the low temperature.
  • the electronic device 100 boosts the output voltage of the battery 142 to avoid abnormal shutdown caused by low temperature.
  • the touch sensor 180K may also be referred to as a touch panel or a touch sensitive surface.
  • the touch sensor 180K may be disposed on the display screen 194 , and the touch sensor 180K and the display screen 194 form a touch screen, also called a “touch screen”.
  • the touch sensor 180K is used to detect a touch operation on or near it.
  • the touch sensor can pass the detected touch operation to the application processor to determine the type of touch event.
  • Visual output related to touch operations may be provided through display screen 194 .
  • the touch sensor 180K may also be disposed on the surface of the electronic device 100 , which is different from the location where the display screen 194 is located.
  • the bone conduction sensor 180M can acquire vibration signals.
  • the bone conduction sensor 180M can acquire the vibration signal of the vibrating bone mass of the human voice.
  • the bone conduction sensor 180M can also contact the pulse of the human body and receive the blood pressure beating signal.
  • the bone conduction sensor 180M can also be disposed in the earphone, combined with the bone conduction earphone.
  • the audio module 170 can analyze the voice signal based on the vibration signal of the vocal vibration bone block obtained by the bone conduction sensor 180M, so as to realize the voice function.
  • the application processor can analyze the heart rate information based on the blood pressure beat signal obtained by the bone conduction sensor 180M, and realize the function of heart rate detection.
  • the keys 190 include a power-on key, a volume key, and the like. Keys 190 may be mechanical keys. It can also be a touch key.
  • the electronic device 100 may receive key inputs and generate key signal inputs related to user settings and function control of the electronic device 100 .
  • Motor 191 can generate vibrating cues.
  • the motor 191 can be used for vibrating alerts for incoming calls, and can also be used for touch vibration feedback.
  • touch operations acting on different applications can correspond to different vibration feedback effects.
  • the motor 191 can also correspond to different vibration feedback effects for touch operations on different areas of the display screen 194 .
  • Different application scenarios for example: time reminder, receiving information, alarm clock, games, etc.
  • the touch vibration feedback effect can also support customization.
  • the indicator 192 can be an indicator light, which can be used to indicate the charging state, the change of power, and can also be used to indicate a message, a missed call, a notification, and the like.
  • the SIM card interface 195 is used to connect a SIM card.
  • the SIM card can be contacted and separated from the electronic device 100 by inserting into the SIM card interface 195 or pulling out from the SIM card interface 195 .
  • the electronic device 100 may support 1 or N SIM card interfaces, where N is a positive integer greater than 1.
  • the SIM card interface 195 can support Nano SIM card, Micro SIM card, SIM card and so on. Multiple cards can be inserted into the same SIM card interface 195 at the same time. The types of the plurality of cards may be the same or different.
  • the SIM card interface 195 can also be compatible with different types of SIM cards.
  • the SIM card interface 195 is also compatible with external memory cards.
  • the electronic device 100 interacts with the network through the SIM card to implement functions such as call and data communication.
  • the electronic device 100 employs an eSIM, ie: an embedded SIM card.
  • the eSIM card can be embedded in the electronic device 100 and cannot be separated from the electronic device 100 .
  • the software system of the electronic device 100 may adopt a layered architecture, an event-driven architecture, a microkernel architecture, a microservice architecture or a cloud architecture.
  • the embodiments of the present application describe the software structure of the electronic device 100 by using a layered architecture.
  • FIG. 3 is a block diagram of a software structure of an electronic device 100 according to an embodiment of the present application.
  • the layered architecture divides the software system into several layers, and each layer has a clear role and division of labor. Layers communicate with each other through software interfaces. In some embodiments, the layered architecture divides the system into four layers, from top to bottom, the application layer, the application framework layer, the Android runtime (Android runtime) and the system library, and the kernel layer.
  • the application layer can include a series of application packages.
  • the application package can include applications such as camera, gallery, calendar, phone, map, memo, address book, weather, music, video, SMS, etc.
  • the application framework layer provides an application programming interface (application programming interface, API) and a programming framework for applications in the application layer.
  • the application framework layer includes some predefined functions.
  • the application framework layer can include a personal knowledge graph manager, a model manager, and so on.
  • the personal knowledge graph manager is used to construct the user's personal knowledge graph.
  • the personal knowledge graph manager includes a knowledge acquisition module, a knowledge fusion module, and a calculation storage module.
  • the knowledge acquisition module is used to acquire personal data.
  • the knowledge acquisition module can acquire the user's personal data from two aspects.
  • the data generated by each application program in the electronic device is stored in the data service and/or the file system, and the knowledge acquisition module can obtain the user's personal data from the data service and/or the file system.
  • the application in the electronic device can obtain the user's authorization, and after obtaining the user's authorization, the electronic device can obtain the user's personal data from each application.
  • the knowledge acquisition module sends the personal data to the knowledge fusion module.
  • the knowledge fusion module is used to receive the personal data sent by the knowledge acquisition module, preprocess the personal data, and classify the preprocessed personal data according to the relationship, event, and entity according to the machine learning algorithm, and obtain the user's relationship knowledge and event knowledge. , entity knowledge these three types of knowledge.
  • User knowledge includes relational knowledge, event knowledge, and entity knowledge.
  • Relationship knowledge refers to user knowledge of interpersonal relationships such as friend relationships, colleague relationships, and family relationships of users obtained from personal data.
  • Event knowledge refers to user knowledge of things that have happened, things that are in progress, or things that have not happened, such as traveling, going on business trips, going to fitness, etc., obtained from personal data.
  • Entity knowledge refers to user knowledge of entities such as movies and music that the user likes obtained from personal data.
  • the personal data acquired by the knowledge acquisition module can be divided into structured personal data and unstructured personal data.
  • the knowledge fusion module can preprocess unstructured personal data (such as audio, video, pictures, etc.). Preprocessing involves converting information contained in unstructured personal data (eg, audio, video, pictures, etc.) into textual information. Then clean the text information. Cleaning the text information includes performing word segmentation and part-of-speech tagging on the text information through natural language processing, and deleting words in the text information that do not belong to the preset part-of-speech according to the preset part-of-speech, such as deleting articles, prepositions, adverbs, conjunctions, verbs , tone words, etc.
  • the preset part of speech can be user-defined, and the preset part of speech can be set according to actual needs.
  • the knowledge fusion module deduplicates the cleaned text information.
  • the purpose of deduplication is that, on the one hand, the electronic device 100 obtains the user's personal data from the data service and/or the file system and obtains the user's personal data directly from various applications, which may cause duplicate data; The device 100 directly obtains the user's personal data from each application program, and the personal data in each application program may also be duplicated. Therefore, the knowledge fusion module deduplicates the cleaned text information, deletes duplicate data, and reduces data redundancy.
  • part-of-speech tagging can refer to Table 1:
  • part-of-speech tagging can be performed on the words after word segmentation of the text information. Specifically, add “/a” after adjectives, “/b” after distinguishing words, “/c” after conjunctions, “/d” after adverbs, and “/d” after interjections "/e”, add “/f” after the position word, add “/g” after the morpheme, add “/h” after the preceding component, add “/i” after the idiom, and add “/i” after the idiom.
  • Table 1 only exemplarily lists some part-of-speech tagging rules, and may also include more part-of-speech tagging rules, and part-of-speech tagging can also refer to other rules, which are not limited in this application.
  • the knowledge fusion module can preprocess the structured personal data, which includes two steps of data cleaning and deduplication.
  • the way of cleaning and deduplication of structured personal data is the same as the way of cleaning and deduplication of unstructured personal data, and will not be repeated here.
  • the machine learning algorithm may be a clustering algorithm based on association rules.
  • Association rules can be preset according to actual needs, and then the preprocessed personal data can be clustered and processed according to the set association rules, and the core information (relationship knowledge, event knowledge and entity knowledge) in the preprocessed personal data can be extracted. Knowledge). This method can more accurately extract the core information (relational knowledge, event knowledge and entity knowledge) related to the constructed knowledge graph in the preprocessed personal data.
  • the core information in the preprocessed personal data (second personal data) unrelated to the constructed knowledge graph is collectively referred to as user basic feature data.
  • the user's basic characteristic data may include: user's gender, age; device information used by the user (eg, device ID, device model), and the like.
  • machine learning algorithms can also be decision tree classification methods, naive Bayesian classification algorithms, and support vector machine-based classification methods. This application is not limited here.
  • the knowledge fusion module is also used for sending relational knowledge, entity knowledge and event knowledge to the computing storage module.
  • the calculation and storage module is used to receive the user knowledge sent by the knowledge fusion module, and store the user knowledge according to a predetermined structure. For example, user knowledge can be stored in a five-tuple structure.
  • the calculation and storage module may store the relational knowledge according to a predetermined structure.
  • the purpose of storing relational knowledge according to a predetermined structure is to store relational knowledge and time correspondingly, reflecting the connection between relational knowledge and time.
  • relationship knowledge and time can be stored according to the structure of the quintuple (the first quintuple structure), and the structure of the quintuple is "entity 1-relation-entity 2-time point-time interval".
  • the calculation and storage module can store the event knowledge according to a predetermined structure.
  • the purpose of storing event knowledge according to a predetermined structure is to store event knowledge and time correspondingly, reflecting the connection between event knowledge and time.
  • event knowledge and time can be stored in a quintuple structure (second quintuple structure), where the quintuple structure is event-argument-logical relationship-time point-time interval.
  • the argument is the action that supports the event, and the logical relationship can be a causal relationship, a succession relationship, and so on.
  • the calculation and storage module can store the entity knowledge according to a predetermined structure. Storing entity knowledge according to a predetermined structure is to store entity knowledge corresponding to time, which reflects the connection between entity knowledge and time. For example, entity knowledge and time can be stored according to a quintuple structure (the third quintuple structure), and the quintuple structure is "entity 1: time-association weight-entity 2-relationship weight-entity 3".
  • TV Series 1 2020.4.6-1.0-Actor1-0.8-Actor3.
  • the entity knowledge expresses that the release time of "TV Series 1" is April 6, 2020, the degree of correlation between "TV Series 1” and Actor 1 is 1.0, and the degree of correlation between Actor 1 and Actor 3 is 0.8. It shows that "TV drama 1" has a higher degree of correlation with actor 1.
  • the computing and storage module is also used for constructing a personal knowledge graph according to the user knowledge of a predetermined structure.
  • Personal knowledge graph is to display user knowledge of predetermined structure in the form of graph.
  • the personal knowledge graph constructed in the form of a graph can be referred to as shown in FIG. 8 .
  • the computing and storage module is also used for sending the user's personal knowledge graph to the personal knowledge graph database for storage.
  • the model manager includes feature pools and model pools.
  • the feature pool is used to store the user's personal knowledge graph and user basic feature data;
  • the model pool is preset with a reordering model, and the reordering model can be logistic regression, decision tree, Factorization Machine (FM), field perception Decomposition machine (Field-aware Factorization Machine, FFM), deep learning and other algorithms.
  • the model pool trains the reordering model based on the user's personal knowledge graph and user basic feature data.
  • the trained re-ranking model (the first model) can further screen the coarsely-ranked advertisements to obtain the advertisements to be placed.
  • the following describes how the model pool trains the reordering model.
  • the input of the training data of the reordering model is historical advertisement information, user basic feature data, and personal knowledge graph, and the output of the training data is the user's historical behavior.
  • the reordering model can be, but is not limited to, logistic regression, decision tree, FFM, deep learning and other algorithms.
  • the model pool stores historical advertisement information of multiple advertisements displayed in the past, and the historical advertisement information may be information such as the ID of the advertisement, the description and size of the advertisement, and the like.
  • User basic feature data may include: user gender, age, device information (eg, device identification, device model) used by the user, and the like.
  • the user's historical behavior may include which advertisements the user has watched in the past week, the duration of the user's viewing of the advertisement, which advertisements the user has closed, and so on.
  • the reranking model will output a result, which can be either the user clicked and viewed the ad or the user did not view and closed the ad. Compare the result with the output of the training data. If the result does not match the output of the training data, modify the parameters of the reordering model and continue to train the reordering model. When the output result of the reordering model conforms to the user's historical behavior, the model training ends.
  • the reordering model can predict the click probability of each advertisement in the coarsely arranged advertisements, that is, to obtain the probability value of each advertisement user clicks, and the reordering model arranges the advertisements according to the probability value of user clicks from large to high. Small to be sorted and get served ads.
  • the user's interpersonal network can be acquired according to the relational knowledge recorded in the user's personal knowledge graph, and whether the user is a social person or an introvert can be learned according to the interpersonal network.
  • the event knowledge recorded in the user's personal knowledge graph the frequently done things of the user can be obtained, such as events such as travel and business trips.
  • the entity knowledge recorded in the user's personal knowledge graph the TV series that the user likes to watch, etc. can be obtained.
  • the re-ranking model learns according to the user's personal knowledge graph that the user is a user who likes to watch movies and TV series and occasionally goes on business trips.
  • Roughly arranged advertisements include a ticket-buying advertisement, a film and television drama advertisement, and a dating advertisement.
  • the re-ranking model predicts the probability of coarsely-arranged advertisements, and obtains that the probability of users clicking on ticket-buying advertisements is 0.5, the probability of users clicking on film and television advertisements is 0.9, and the probability of users clicking on dating advertisements is 0.4.
  • the reordering model reorders the coarsely arranged advertisements according to the probability of user clicks, and obtains the advertisements.
  • the order in which advertisements are placed is film and television drama advertisements, ticket purchase advertisements, and friendship advertisements.
  • the application program obtains the reordering model, and the application program can screen the coarsely-arranged advertisements according to the trained reordering model (the first model) to obtain the advertisements to be placed.
  • the advertisement placement is obtained by re-screening the advertisements in the coarsely arranged advertisements according to the user's personal behavioral characteristics, and the placement of advertisements is obtained by sorting the advertisements according to the probability that the user may click and watch the advertisements from high to low.
  • the following describes how the application program filters the coarsely arranged advertisements to obtain the advertisements.
  • the application program after the application program obtains the coarsely arranged advertisements, the application program obtains the reordering model, and according to the reordering model, sorts the coarsely arranged advertisements according to the probability of the advertisements that the user may click and watch from high to low, and obtains the placement. advertise.
  • the application program after the application program obtains the rough-arranged advertisements, the application program obtains the re-ranking model, and the application program may only retain the advertisement with the highest probability that the user may click and view the advertisement personally.
  • the application program after the application program obtains the coarsely arranged advertisements, the application program obtains the reordering model, and the application program can filter out the advertisements whose probability that the user may click and view the advertisements is lower than the threshold.
  • the application program after the application program obtains the coarsely arranged advertisements, the application program obtains the reordering model, and according to the reordering model, sorts the coarsely arranged advertisements according to the probability of the advertisements that the user may click and watch from high to low, and the application The program will also determine whether each advertisement has been pushed within a certain period of time (such as three days). If the application program determines that some advertisements in each advertisement have been pushed within a certain period of time (such as three days), the application program will filter out the Advertisements that have been pushed within a certain period of time (for example, three days).
  • the model manager in the application layer can provide an interface.
  • the application server sends an advertisement recommendation request to the ad server. After that, the application receives the rough advertisement sent by the ad server.
  • the application can obtain the user's authorization through the model manager.
  • the interface obtains the reordering model, and the application filters the coarsely arranged advertisements through the reordering model to obtain the advertisements.
  • the application recommends the obtained advertisement to the user for viewing.
  • the personal knowledge graph library in the system library can provide an interface, and after the application program obtains the authorization of the user, the application program can obtain the user's personal knowledge graph from the interface.
  • the application can recommend the content of interest to the user through the user's personal knowledge graph.
  • the newly downloaded application has not yet recorded the user's behavior, and the newly downloaded application may obtain the user's authorization to obtain the user's personal knowledge graph.
  • the newly downloaded application can recommend content of interest to the user through the user's personal knowledge graph.
  • users are not required to select the content they are interested in, and newly downloaded applications can make personalized recommendations based on personal knowledge graphs.
  • newly downloaded applications can make personalized recommendations based on personal knowledge graphs, and the recommended content conforms to the user's behavioral characteristics, optimizing the content recommendation effect.
  • Android Runtime includes core libraries and a virtual machine. Android runtime is responsible for system scheduling and management.
  • the core library consists of two parts: one is the functional functions that the java language needs to call, and the other is the core library of the system.
  • the application layer and the application framework layer run in virtual machines.
  • the virtual machine executes the java files of the application layer and the application framework layer as binary files.
  • the virtual machine is used to perform functions such as object lifecycle management, stack management, thread management, safety and exception management, and garbage collection.
  • a system library can include multiple functional modules. For example: data service, file system, personal knowledge map library, surface manager (surface manager), 3D graphics processing library (eg: OpenGL ES), etc.
  • the data service is used to store structured data involving user privacy, such as databases, tables, etc., generated during the process of running and storing application programs on electronic devices.
  • the file system is used to store unstructured data related to user privacy, such as documents, pictures, videos and other data generated during the process of running and storing application programs on the electronic device.
  • the personal knowledge graph database is used to store personal knowledge graphs.
  • the Surface Manager is used to manage the display subsystem and provides a fusion of 2D and 3D layers for multiple applications.
  • the 3D graphics processing library is used to implement 3D graphics drawing, image rendering, compositing, and layer processing.
  • the kernel layer is the layer between hardware and software.
  • the kernel layer contains at least display driver, camera driver, audio driver, and sensor driver.
  • FIG. 4 is a schematic diagram of a hardware structure of an advertisement server 200 provided by an embodiment of the present application.
  • Ad server 200 may include: one or more processors 301 , memory 302 , communication interface 303 , transmitter 305 , receiver 306 , coupler 307 , and antenna 308 . These components may be connected through a bus 304 or other means, and FIG. 3 takes the connection through a bus as an example. in:
  • the communication interface 303 can be used for the advertisement server 200 and other communication devices, and the other communication devices can be, for example, the above-mentioned electronic devices or other network devices.
  • the application server sends an advertisement recommendation request to the advertisement server 200
  • the advertisement server 200 receives and responds to the advertisement recommendation request sent by the application server
  • the advertisement server 200 sends coarsely arranged advertisements to the application in the electronic device 100 .
  • the communication interface 303 may be a Long Term Evolution (LTE) (4G) communication interface.
  • LTE Long Term Evolution
  • the advertisement server 200 may also be configured with a wired communication interface 303 to support wired communication, for example, the backhaul link between the advertisement server 200 and other communication devices may be a wired communication connection.
  • the transmitter 305 and the receiver 306 may be regarded as a wireless modem.
  • the transmitter 305 can be used for transmitting and processing the signal output by the processor 301 .
  • a receiver 306 may be used to receive signals.
  • the number of transmitters 305 and receivers 306 may be one or more.
  • Antenna 308 may be used to convert electromagnetic energy in a transmission line to electromagnetic waves in free space, or to convert electromagnetic waves in free space to electromagnetic energy in a transmission line.
  • the coupler 307 can be used to divide the mobile communication signal into multiple paths for distribution to a plurality of receivers 306 .
  • Memory 302 is coupled to processor 301 for storing various software programs and/or sets of instructions.
  • memory 302 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic disk storage devices, flash memory devices, or other non-volatile solid-state storage devices.
  • the memory 302 may store an operating system (hereinafter referred to as a system), such as an embedded operating system such as uCOS, VxWorks, and RTLinux.
  • a system such as an embedded operating system such as uCOS, VxWorks, and RTLinux.
  • the memory 302 may also store a network communication program that can be used to communicate with one or more additional devices, one or more terminal devices, and one or more network devices.
  • the processor 301 may be configured to read and execute computer-readable instructions. Specifically, the processor 301 may be configured to call a program stored in the memory 302, such as an implementation program of the method provided by one or more embodiments of the present application on the side of the advertisement server 200, and execute the instructions contained in the program.
  • a program stored in the memory 302 such as an implementation program of the method provided by one or more embodiments of the present application on the side of the advertisement server 200, and execute the instructions contained in the program.
  • the hardware structure of the advertisement server 200 shown in FIG. 4 is only an implementation manner of the embodiments of the present application. In practical applications, the advertisement server 200 may further include more or less components, which is not limited here.
  • FIG. 5 is a schematic diagram of the architecture of another advertisement recommendation system provided by an embodiment of the present application.
  • the electronic device 100 may include a knowledge acquisition module 5101 , a knowledge fusion module 5102 , a calculation and storage module 5103 , and a reordering module 5104 .
  • each module of the knowledge acquisition module 5101, the knowledge fusion module 5102, the calculation storage module 5103, and the reordering module 5104 please refer to the embodiment shown in FIG.
  • the advertisement server 200 may include: an advertisement click module 5201 , a dynamic advertisement pool module 5202 , a rough advertisement arrangement module 5203 , and a group portrait module 5204 .
  • the advertisement click module 5201 can be configured to receive an advertisement recommendation request sent by the application server. In response to the advertisement recommendation request sent by the application server, the advertisement click module 5201 sends an advertisement bidding request to the dynamic advertisement pool module 5202.
  • the dynamic advertisement pool module 5202 can be configured to receive and respond to the advertisement bidding request sent by the advertisement click module 5201 .
  • the dynamic advertisement pool module 5202 may send a request to obtain advertisements to multiple advertiser servers, the multiple advertiser servers receive and respond to the obtain advertisement requests, and the multiple advertiser servers send advertisements to the dynamic advertisement pool module 5202.
  • the dynamic advertisement pool module 5202 filters the advertisements sent by multiple advertiser servers (for example, according to the price of advertisements from high to low) to obtain an advertisement set, and the dynamic advertisement pool module 5202 sends the advertisement set to the rough advertisement ranking Module 5203.
  • the advertisement rough arrangement module 5203 can be used to receive the advertisement set sent by the dynamic advertisement pool module 5202.
  • the advertisement rough arrangement module 5203 is also used to receive the group portrait sent by the group portrait module 5204.
  • the advertisement rough arrangement module 5203 according to the group portrait. Screening is performed to obtain coarse-arranged advertisements, which may include one or more advertisements, and the coarse-arranged advertisements module 5203 sends the coarse-arranged advertisements to the application program.
  • the application program in the electronic device 100 receives the rough-arranged advertisement sent by the advertisement rough-arrangement module 5203, and obtains a reordering model.
  • the application filters the coarsely arranged advertisements according to the reordering model, and obtains the advertisements to be placed.
  • the application recommends serving advertisements to users for viewing.
  • the group portrait module 5204 can be used to obtain the group data of the user group sent by the application server, then the group data can be which advertisements the user group clicked, which advertisements the user group viewed, which advertisements were closed by the user group, and which advertisements were viewed by the user group. duration etc.
  • the group portrait module 5204 performs group portrait for the user group according to the group data of the user group.
  • the group portrait is to label the user group according to the group data of the user group. A type of advertisement, etc.
  • the following describes how the electronic device directly obtains the user's personal data from the application program in conjunction with the user interface on the electronic device.
  • FIG. 6A exemplarily shows an exemplary user interface 60 of an address book application on an electronic device.
  • User interface 40 may include a status bar 400 and an address book list 410 .
  • User contacts can be classified according to attributes such as friends, colleagues, relatives, etc.
  • the contacts of the friend category may include Wang Ke, whose phone number is 12345; the contacts of the friend category also include Li Ke, whose phone number is 23456.
  • Colleague contacts can include Zhang San, whose phone number is 12346; Colleague contacts also include Lao Wang, whose phone number is 12045; Colleague contacts can include Zhang San, Zhang San's phone number It is 12365; the contacts of colleagues also include Xiao Li, and Xiao Li's phone number is 23045; the contacts of colleagues also include Huahua, whose phone number is 92345.
  • Contacts for the loved one can include Mom, whose phone number is 65430.
  • the personal data generated in the process of running the short message application on the electronic device is introduced by way of example in conjunction with the user interface on the electronic device.
  • the short message application records short message information related to the user, and the electronic device can obtain behavior information related to the user from the short message information. For example, if a user purchases a ticket from Beijing to Shanghai through a ticketing application, the ticketing application sends the user's ticket-purchasing SMS to the SMS application, and the electronic device can obtain the places where the user frequently goes from the SMS application. For another example, if the user purchases a movie ticket through the application, the application sends the user's ticket-purchasing short message to the short message application, and the electronic device can obtain the type of movie the user likes to watch from the short message application.
  • the electronic device can also directly obtain the user's ticket-purchasing short message from the ticket-purchasing application. This application is not limited here.
  • FIG. 6B exemplarily shows an exemplary user interface 50 of a short message application on an electronic device.
  • the user interface 50 may include a status bar 400 and a text message content display box 420 .
  • the text message content display box 420 may include the text message content, and the text message content is "Order EK123456, on August 24, 2020, Z234, No. 10, No. 24, bottom bunk, Shenzhen Station opens at 12:45, and the arrival station is Beijing West. Please hold the ticket certificate. Enter the station and board the train. You have purchased an e-ticket, please go to the ticket check and wait directly with the ticket purchase certificate.”
  • the electronic device can know from the content of the text message that the user will take a train from Shenzhen Railway Station to Beijing West Railway Station on August 24, 2020, so the electronic device recommends hotel accommodation advertisements to the user for viewing.
  • the following describes personal data generated during the process of running communication applications on the electronic device by way of example in conjunction with the user interface on the electronic device.
  • FIG. 6C illustrates an exemplary user interface of a communication-type application on an electronic device.
  • the communication application may be a phone application preset on the electronic device.
  • the communication application can also be a communication application downloaded by the user from the application store, and the user can make a voice call or video call with a friend through the communication application. This application is not limited here.
  • the communication application is a phone application program preset by the electronic device.
  • the electronic device runs the phone application, it can obtain the call frequency between the user's contact and the user according to the user's call record, so that the electronic device can infer the intimacy of the user's contact with the user according to the frequency of the user's contact and the user.
  • FIG. 6C exemplarily shows an example user interface 40 of a telephony application on an electronic device.
  • User interface 40 may include a status bar 400 and a call log list 410 .
  • Status bar 400 may include a time indicator 4001, a battery status indicator 4002, one or more signal strength indicators 4003 for wireless fidelity (Wi-Fi) signals, a mobile communication signal (also referred to as a cellular signal) One or more signal strength indicators 4004 of the .
  • Wi-Fi wireless fidelity
  • a mobile communication signal also referred to as a cellular signal
  • the call record list 410 may include one or more call records. For example, at 10:22 this morning, the user made a call with Zhang San, and at 11:30 this morning, the user made a call with Lao Wang, At 1:15 this afternoon, the user made a call with Huahua, at 1:21 this afternoon, the user made a call with his mother, and at 2:11 this afternoon, the user made a call with Mr. Wang, At 6:10 this afternoon, the user made a call with Wang Ke, at 6:10 this afternoon, the user made a call with Li Ke, and at 8:15 this evening, the user made a call with his mother. Yesterday, the user made a call with Li Ke. Had a phone call with my mom.
  • the call record list 410 may also include more or less call records, which is not an example here.
  • the following describes personal data generated during the process of running the memo application on the electronic device by way of example in conjunction with the user interface on the electronic device.
  • FIG. 6D exemplarily shows an example user interface 70 of a memo application on an electronic device.
  • User interface 40 may include status bar 400 and my memo list 440 .
  • My memo list 440 may include important things, important people, and times noted by the user. For example, “work report at 2:00 pm on August 7th, prepare PPT”, “Wang Ke's birthday on August 16th, remember to buy birthday gifts”, “flight at 3:00 pm on August 20th, travel to Beijing on business”. My memo list 440 can also record more or less important things, important people and times, which will not be repeated here.
  • the following describes personal data generated during the process of running the memo application on the electronic device by way of example in conjunction with the user interface on the electronic device.
  • the weather application records the weather conditions of the user-preset cities.
  • the electronic device may acquire the city where the user is located, the weather conditions of the city, and the like from the weather module 2106 .
  • FIG. 6E exemplarily shows an example user interface 80 for weather on an electronic device.
  • User interface 80 may include a status bar 400 and a list 450 of weather conditions for the week.
  • the list of weather conditions of the week 450 may include daily weather conditions for a location during the week.
  • the location of the electronic equipment is Nanshan District, Shenzhen.
  • the weather at that time (Tuesday) was cloudy and the temperature was 29°.
  • Today (Tuesday) was cloudy and the temperature was 28° at 9:00 pm.
  • Today (Tuesday) The weather at ten o'clock in the evening is cloudy with a temperature of 28°
  • today (Tuesday) at eleven o'clock in the evening is cloudy with a temperature of 27°
  • tomorrow (Wednesday) at 0 a.m. will be cloudy with a temperature of 27°
  • the above UI embodiment is just an example to illustrate that the electronic device 100 can obtain the user's personal data from the above application program.
  • Personal data can also come from other applications, which are not introduced here in this application.
  • FIG. 7 is a schematic flowchart of an advertisement display method provided by an embodiment of the present application.
  • the method includes:
  • the electronic device 100 acquires personal data, and preprocesses the personal data.
  • the data generated by each application program in the electronic device is stored in the data service and/or the file system, and the electronic device 100 can obtain the user's personal data from the data service and/or the file system.
  • the application in the electronic device can obtain the user's authorization, and after obtaining the user's authorization, the electronic device can directly obtain the user's personal data from each application.
  • the electronic device 100 may directly acquire the user's interpersonal relationship knowledge, such as the user's friend relationship, the user's colleague relationship, the user's relatives, the user's customer relationship, and other user knowledge from an address book application, a communication application, or the like.
  • the user's interpersonal relationship knowledge such as the user's friend relationship, the user's colleague relationship, the user's relatives, the user's customer relationship, and other user knowledge from an address book application, a communication application, or the like.
  • the electronic device 100 can directly acquire user knowledge such as the user's friend's birthday, conference information, travel city, academic papers, departure date, etc. from the short message application, the memo application, and the like.
  • the electronic device 100 may acquire user knowledge such as the user's favorite things, favorite cities, favorite singers, song titles, and interested actors from the gallery application.
  • the electronic device 100 may acquire user knowledge such as the city area where the user is located, weather conditions, and exercise status from the wearable device (eg, a Bluetooth watch).
  • the wearable device eg, a Bluetooth watch
  • the electronic device 100 obtains description information and data of "TV Drama 1" from a video application.
  • the description information and data of "TV Series 1" are shown in Table 2:
  • the electronic device 100 is also used for preprocessing from personal data.
  • Preprocessing includes cleaning and deduplication of personal data.
  • the cleaning of personal data includes word segmentation and part-of-speech tagging of text information through natural language processing, and deletion of words in the text information that do not belong to the preset part-of-speech according to the preset part-of-speech, such as deleting articles, prepositions, adverbs, conjunctions, verbs , tone words, etc.
  • the preset part of speech can be user-defined, and the preset part of speech can be set according to actual needs.
  • the electronic device 100 obtains the user's personal data from the data service and/or the file system and directly obtains the user's personal data from various applications, there will be duplicate data; on the other hand, the electronic device 100 directly obtains the user's personal data from Each application directly obtains the user's personal data, and the personal data in each application may also be duplicated. Therefore, the knowledge fusion module deduplicates the cleaned text information, deletes duplicate data, and reduces data redundancy.
  • the electronic device 100 preprocesses the acquired description information and data of "TV Drama 1", which specifically includes performing word segmentation and part-of-speech tagging on the description information and data of "TV Drama 1".
  • TV Drama 1 which specifically includes performing word segmentation and part-of-speech tagging on the description information and data of "TV Drama 1”.
  • part-of-speech tagging refer to the embodiment in FIG. 3 , which is not repeated in this application.
  • the electronic device 100 cleans the description information and data of "TV Drama 1" after word segmentation and part-of-speech tagging according to a preset part-of-speech, and obtains words including person names, place names, song titles, film and television drama titles, time, actors, etc.
  • the description information and data of "TV Series 1" after word segmentation and part-of-speech tagging are shown in Table 4:
  • the electronic device 100 classifies the preprocessed personal data according to the relationship, event, and entity according to the machine learning algorithm, and obtains three types of knowledge of the user: relationship knowledge, event knowledge, and entity knowledge.
  • the electronic device 100 stores the user knowledge according to a predetermined structure, and constructs a personal knowledge graph according to the user knowledge of the predetermined structure.
  • the electronic device 100 stores the relational knowledge according to a predetermined structure.
  • the purpose of storing relational knowledge according to a predetermined structure is to store relational knowledge and time correspondingly, reflecting the connection between relational knowledge and time.
  • relationship knowledge and time can be stored according to the structure of five-tuple, and the structure of five-tuple is "entity 1-relation-entity 2-time point-time interval".
  • time interval is represented by the number of months and 203 days, it may also be represented by the number of years, and may also be represented by the number of days, which is not limited in this application.
  • the electronic device 100 may also only represent the relationship between entities as only time points.
  • the basic format is: Entity 1 - Relationship - Entity 2 - Time Point.
  • the electronic device 100 stores the event knowledge according to a predetermined structure.
  • the purpose of storing event knowledge according to a predetermined structure is to store event knowledge and time correspondingly, reflecting the connection between event knowledge and time.
  • event knowledge and time can be stored according to the five-tuple structure, and the five-tuple structure is "event-argument-logical relationship-time point-time interval".
  • the argument is the action that supports the event
  • the logical relationship can be a causal relationship, a succession relationship, and so on.
  • Exemplary go to duty-free shop - buy cosmetics - cause and effect - 2020.8.20 - 2 days.
  • the knowledge of the event expresses the purchase event. Going to the duty-free shop and buying cosmetics is a causal relationship.
  • the departure date is August 20, 2020, and the itinerary is two days.
  • the electronic device 100 may also only represent events that occur to the user as time points.
  • the basic format is: entity 1: event-argument-logical relationship-time point.
  • the electronic device 100 stores the entity knowledge according to a predetermined structure. Storing entity knowledge according to a predetermined structure is to store entity knowledge corresponding to time, which reflects the connection between entity knowledge and time. For example, entity knowledge and time can be stored according to a five-tuple structure, and the five-tuple structure is "entity 1: time-association weight-entity 2-relationship weight-entity 3".
  • TV Series 1 2020.4.6-1.0-Actor1-0.8-Actor3.
  • the entity knowledge expresses that the release time of "TV Series 1" is April 6, 2020, the degree of correlation between "TV Series 1” and Actor 1 is 1.0, and the degree of correlation between Actor 1 and Actor 3 is 0.8. It shows that "TV drama 1" has a higher degree of correlation with actor 1.
  • TV Drama 2 2020.2.1-0.6-actor2-0.8-actor1.
  • the entity knowledge expresses that the release time of "TV Series 2" is February 1, 2020, the degree of correlation between "TV Series 2" and Actor 2 is 0.6, and the degree of correlation between Actor 2 and Actor 1 is 0.8. It shows that "TV drama 2" has a higher degree of correlation with actor 1.
  • the electronic device 100 performs entity mining on the cleaned and deduplicated description information and data of TV Drama 1, and establishes time-based entity knowledge, TV Drama 1: 2020-1.0-Actor1-0.8-Actor2.
  • the electronic device 100 constructs a personal knowledge graph according to the event knowledge of the predetermined structure, the relational knowledge of the predetermined structure, and the entity knowledge of the predetermined structure.
  • Personal knowledge graph is to represent user knowledge of predetermined structure in the form of graph.
  • FIG. 8 is a schematic diagram showing the constructed personal knowledge graph in the form of a graph in an embodiment of the present application. It can be understood that the personal knowledge graph in FIG. 8 only shows the personal knowledge graph of the user's partial user knowledge, and the personal knowledge graph may also include more or less user knowledge.
  • the user's personal knowledge graph may include relational knowledge, event knowledge and entity knowledge.
  • the electronic device 100 since the user knowledge acquired by the electronic device 100 from each application program module changes in real time, the electronic device 100 also needs to update the personal knowledge graph in real time according to the user knowledge acquired from each application program module. In this way, the personal knowledge graph will more accurately express the user's characteristics.
  • the electronic device 100 may further update the user knowledge in the personal knowledge graph.
  • the specific personal knowledge graph may only retain the user knowledge of the user in the most recent period, and the electronic device 100 may filter out the user knowledge that does not belong to the most recent period in the personal knowledge graph.
  • the user characteristics depicted in the updated personal knowledge graph are more acurrate.
  • the storage resources of the electronic device 100 can be saved.
  • the following describes how the electronic device 100 updates the personal knowledge graph.
  • the update of the personal knowledge graph by the electronic device 100 can be divided into two aspects.
  • the following describes how the electronic device 100 filters out the existing knowledge of the user in the personal knowledge graph.
  • the electronic device 100 may only retain user knowledge in the personal knowledge graph for a recent period (for example, two years), and filter out user knowledge in the personal knowledge graph beyond the recent period (for example, two years). Lose.
  • the electronic device 100 may also filter out the knowledge that the user already exists in the personal knowledge graph according to the memory size of the personal knowledge graph.
  • the electronic device 100 may retain knowledge in the personal knowledge graph for a recent period according to the preset memory size.
  • the embodiments of the present application may also update the user's personal knowledge graph in other ways.
  • the above embodiments are only used to explain the present application and should not be construed as limitations.
  • the electronic device 100 may update the personal knowledge graph in a specific time interval.
  • the electronic device 100 may update the personal knowledge graph at a fixed time (eg, one day).
  • the electronic device 100 may update the personal knowledge graph according to the user's behavior habits. Exemplarily, during the time period "22:00-8:00", the user rests at home, the electronic device 100 is in a standby state, and the electronic device 100 can update the personal information during the time period "22:00-8:00". Knowledge Graph. Because when the electronic device 100 updates the personal knowledge graph, a certain memory space of the electronic device 100 will be occupied. In this way, the electronic device 100 does not update the personal knowledge graph during the user's rest time, leaving more memory space for the user's operation, and ensuring the smoothness of the user's operation.
  • the electronic device 100 trains the re-ranking model according to historical advertisement information, basic user feature data, personal knowledge graph, and user historical behavior.
  • the electronic device 100 trains the reordering model based on data such as personal knowledge graphs. From the above embodiment, it can be known that the personal knowledge graph will be updated over time, so the electronic device 100 should also be updated over time. Update the re-ranking model so that the ranking list of advertisements obtained by the re-ranking model re-ranking multiple input advertisements is more accurate.
  • the following describes how the electronic device 100 updates the reordering model row.
  • the electronic device 100 may update the reordering model at a fixed time (eg, one day).
  • the electronic device 100 may update the personal knowledge graph according to the user's behavioral habits.
  • the electronic device 100 is in a standby state, and the electronic device 100 can update the reload during the time period "22:00-8:00". Sort the model. Because the electronic device 100 will occupy a certain memory space of the electronic device 100 when updating the reordering model, the electronic device 100 will not update the reordering model while the user is using the electronic device 100, leaving more memory space for the user to operate, ensuring that the user Smoothness of operation.
  • the application program in the electronic device 100 receives the rough layout advertisement sent by the advertisement server 200,
  • the electronic device 100 detects the user operation, and in response to the user operation, the electronic device 100 sends a user request to the application server 300 , and the application server 300 sends an advertisement recommendation request to the advertisement server 200 .
  • the user operation may be: the electronic device 100 detects that the user has opened the application, or the electronic device 100 detects that the user next video is the current browsing page of the application (for example, the electronic device 100 detects that the user swipes down with one finger to enter the next page) video current browsing interface), the electronic device 100 sends a user request to the application server 300 .
  • FIG. 9A exemplarily shows an exemplary user interface 90 of the current browsing interface of the application.
  • the user interface 90 may include a status bar 400 , a current video browsing interface 460 , and an advertisement display area 470 .
  • the current video browsing interface 460 includes a recommendation control 4601 , a popular control 4602 , a small video control 4603 , a next video control 4604 , a video display window 4605 , a like control 4606 , a comment control 4607 , and a forward control 4608 .
  • the advertisement display area 470 includes an advertisement icon 4609 , a close control 4610 , and an advertisement link control 4611 .
  • the video display window 4605 displays the video content 4600 of the currently playing video (eg TV drama 1).
  • the next video control 4604 can receive a user's click operation, and in response to the user's click operation, the current video browsing interface 460 will display the video content of other videos.
  • the electronic device 100 sends a user request to the application server 300, the application server 300 sends an advertisement recommendation request to the advertisement server 200, and in response to the advertisement recommendation request, the advertisement server 200 will send a new The advertisement is sent to the application in the electronic device 100, and the advertisement display area 470 will display the new advertisement content.
  • the current browsing interface of the application can also accept the user's single-finger sliding down to the current browsing interface of the next video, and in response to the user's single-finger sliding operation acting on the current browsing interface, the electronic device 100 sends a user request to the application server 300, The application server 300 sends an advertisement recommendation request to the advertisement server 200. In response to the advertisement recommendation request, the advertisement server 200 will send a new advertisement to the application in the electronic device 100, and the advertisement display area 470 will display the new advertisement content.
  • the following describes how the advertisement server 200 sends the rough advertisement to the application program in the electronic device 100 .
  • the advertisement server 200 performs a group portrait for the user group according to the group data.
  • the advertisement server 200 obtains group data, and the group data may be which advertisements the user clicked, which advertisements the user browsed, which advertisements the user closed, and the like.
  • 9B-9C illustrate UI diagrams of exemplary user interfaces for the ad server 200 to obtain community data.
  • FIG. 9B exemplarily shows an exemplary user interface 900 of the current browsing interface of the application.
  • the user interface 900 may include a status bar 400 , a current video browsing interface 460 , and an advertisement display area 470 .
  • the current video browsing interface 460 includes a recommendation control 4601 , a popular control 4602 , a small video control 4603 , a next video control 4604 , a video display window 4605 , a like control 4606 , a comment control 4607 , and a forward control 4608 .
  • the advertisement display area 470 includes an advertisement icon 4609 , a close control 4610 , and an advertisement link control 4611 .
  • the video display window 4605 displays the video content 4600 of the currently playing video (eg TV drama 1).
  • the next video control 4604 can receive a user's click operation, and in response to the user's click operation, the current video browsing interface 460 will display the video content of other videos.
  • Advertisement content is displayed in the advertisement display area 470, and the advertisement content is "The way to receive coupons has been placed below, please click to receive ABCDEFGHI".
  • the advertisement link control 4611 can receive a user's click operation, and in response to the user's click operation, the user interface 910 will display the user interface of the advertisement. Meanwhile, in response to the user's operation of clicking the advertisement link control 4611, the application server 300 reports the user's behavior of clicking and viewing this advertisement to the advertisement server 200.
  • the behavior of the user clicking and viewing the advertisement can be regarded as a piece of business data of the advertisement.
  • the closing control 4610 can also receive a user's click operation.
  • the advertisement display area 470 displays a selection prompt box 4612 as shown in FIG. 9C . Block this type of advertising controls 4615.
  • the disinterested control 4613 can receive the user's click operation, and in response to the user's click operation, the advertisement display area 470 will no longer display the advertisement; the repeat recommendation control 4614 can receive the user's click operation, and in response to the user's click operation, The advertisement display area 470 will not recommend this advertisement to the user for viewing within a certain period of time (for example, 48 hours); the control 46154614 to block this type of advertisement can receive the user's click operation, and in response to the user's click operation, the advertisement display area 470 will not recommend it.
  • This type of advertisement (for example, beauty cosmetics) is displayed to users.
  • the application server 300 reports to the advertisement server 200 the behavior that the user closes and does not view the advertisement.
  • the user's behavior of closing and not viewing the advertisement can be used as a group data of the advertisement.
  • the group data of the advertisement may also be derived from other channels.
  • the above embodiments are only used to explain the present application, and the embodiments of the present application are not limited herein.
  • the application server 300 collects group data and reports it to the advertisement server 200 .
  • the advertisement server 200 performs group portraits for the user groups according to the groups of all users, and the group portraits label the user groups according to the group data of all users, such as which type of advertisement the user group likes to watch, and which type of advertisement the user group does not like to watch. advertisements, etc.
  • the group data of all users collected by the application server 300 changes in real time, so the group data of all users that the advertisement server 200 receives and sent by the application server 300 also changes in real time.
  • the advertisement server 200 may perform group portraits for the user groups at a fixed time (eg, one day) according to the group data of all users. Then the group portrait of the user group is also updated periodically.
  • the advertisement server 200 screens the advertisements according to the group portraits of the user groups, and obtains coarsely arranged advertisements.
  • a rough placement ad can be one or more ads.
  • Roughly arranged advertisements are obtained by screening multiple advertisements in the advertisement set according to the group portrait of the user group.
  • the advertisement server 200 transmits the rough advertisement to the application in the electronic device 100 .
  • the application program in the electronic device 100 receives the rough layout advertisement sent by the advertisement server 200 .
  • the advertisement server 200 may also classify user groups, for example, the user groups may be divided into female user groups and male user groups, or the user groups may also be divided into users of various age groups. groups, etc.
  • the advertisement server 200 when the advertisement server 200 divides the user group into a female user group and a male user group, the advertisement server 200 profiles the male user group and the female user group respectively.
  • the advertisement server 200 screens a plurality of advertisements according to the group portraits of the male users, and obtains rough-arranged advertisements of the male users.
  • the advertisement server 200 pushes the rough advertisement of the male user to the electronic device of the male user.
  • the advertisement server 200 screens a plurality of advertisements according to the group portraits of the female users, and obtains coarsely arranged advertisements of the female users.
  • the advertisement server 200 pushes the rough advertisement of the female user to the application program in the electronic device of the female user.
  • the group portraits may also profile the user groups according to the user groups of various age groups.
  • the advertisement server 200 draws a portrait of a group of users whose age is between 21-35 years old, and filters a plurality of advertisements according to the group portrait of users between the age of 21-35 years old, and obtains a group of users between the ages of 21-35 Rough ads for users between the ages.
  • the advertisement server 200 pushes the rough-arranged advertisements of users between the ages of 21-35 to the applications in the electronic devices of the users between the ages of 21-35.
  • the advertisement server 200 draws a portrait of a user group whose age is between 36 and 50, and screens multiple advertisements according to the group portrait of the user between the age of 36 and 50, and obtains an age between 36 and 50. of users' rough placement ads.
  • the advertisement server 200 pushes the rough advertisements of the users aged between 36-50 to the application programs in the electronic devices of the users aged between 36-50.
  • the advertisement server 200 draws a portrait of a group of users whose age is between 51 and 70, and filters a plurality of advertisements according to the group portrait of the user between the ages of 51 and 70, and obtains an age between 51 and 70. of users' rough placement ads.
  • the advertisement server 200 pushes the rough-arranged advertisements of users aged between 51-70 to the applications in the electronic devices of the users aged between 51-70.
  • the application program in the electronic device 100 obtains the reordering model, and the application program screens the coarsely arranged advertisements according to the reordering model to obtain the advertisements to be placed.
  • the number of ads served can be one or more ads.
  • the electronic device 100 starts an application program, and the current browsing interface of the application program has an advertisement space, and the advertisement space can display one or more advertisements.
  • the advertisement duration of the advertisement slot is 60 seconds, then the advertisement slot may only display one advertisement, and the duration of the advertisement is 60 seconds.
  • the advertisement slot can display 6 advertisements, and the display time of each advertisement is 10 seconds.
  • Content recommendation may include song recommendation, e-book recommendation, movie and TV drama recommendation, food recommendation, shopping recommendation, etc., which is not limited in this application.
  • FIG. 10 is a schematic flowchart of another advertisement display method provided by an embodiment of the present application.
  • the method includes:
  • the electronic device 100 acquires first personal data.
  • the electronic device 100 acquires the first personal data, and the first personal data is the personal information of the user.
  • Personal information can be one or more of the following: gender, age, personality, hobbies, interpersonal relationships, income, address book information, call records, text messages, memo information, address where you live, and weather conditions at the address where you live.
  • the electronic device 100 may acquire the user's first personal data every fixed period (eg, one week).
  • the electronic device 100 to obtain the first personal data please refer to the embodiments of FIGS. 6A-6D and the embodiment described in S701 , which will not be repeated in this application.
  • the electronic device 100 constructs a personal knowledge graph according to the first personal data.
  • the electronic device 100 Before constructing a personal knowledge graph according to the first personal data, the electronic device 100 needs to preprocess the first personal data.
  • the preprocessing of the first personal data by the electronic device 100 includes the following two steps:
  • Step 1 The electronic device converts the first personal data into text information, and performs sentence segmentation, word segmentation and part-of-speech tagging on the text information.
  • the electronic device acquires words belonging to a preset part of speech from the text information.
  • Step 2 After the electronic device obtains the words belonging to the preset part of speech from the text information, the electronic device 100 deduplicates the words in the text information to remove data redundancy.
  • the electronic device 100 needs to obtain words that appear once in the text information; when two or more identical words appear in the text information, the electronic device 100 retains two or more identical words in the text information a word in .
  • the electronic device 100 acquires the second personal data from the first personal data.
  • the second personal data includes relational knowledge, event knowledge and entity knowledge.
  • the electronic device 100 constructs a personal knowledge graph according to the first personal data, and specifically includes the following steps:
  • the electronic device 100 stores relational knowledge, event knowledge and entity knowledge in a predetermined structure.
  • the predetermined structure may be a quintuple structure.
  • the electronic device 100 stores the relationship knowledge according to the first quintuple structure;
  • the first quintuple structure is "first entity-relationship-second entity-first time point-first time interval";
  • the relationship represents The relationship between the first entity and the second entity, the first time point is the time when the first entity establishes the relationship with the second entity, and the first time interval is the interval from the first time point to the current time point.
  • the electronic device 100 stores the event knowledge according to the second quintuple structure; the second quintuple structure is "event-argument-logical relationship-second time point-second time interval"; the argument is the occurrence action of the event , the logical relationship represents the relationship between the event and the argument, the second time point is the time when the event occurs, and the second time interval is the interval from the second time point to the current time point.
  • the second quintuple structure is "event-argument-logical relationship-second time point-second time interval"
  • the argument is the occurrence action of the event
  • the logical relationship represents the relationship between the event and the argument
  • the second time point is the time when the event occurs
  • the second time interval is the interval from the second time point to the current time point.
  • the electronic device 100 stores the entity knowledge according to the third quintuple structure;
  • the third quintuple structure is "third entity: third time point-first association weight-fourth entity-second association weight-fifth entity ”;
  • the third time point is the occurrence time of the third entity,
  • the first association weight is the association degree between the third entity and the fourth entity, and
  • the second association weight is the association degree between the fourth entity and the fifth entity.
  • the electronic device constructs the user's personal knowledge graph according to the relational knowledge of the predetermined structure, the event knowledge of the predetermined structure, and the entity knowledge of the predetermined structure.
  • the electronic device 100 may also update the personal knowledge graph.
  • the electronic device 100 can delete personal data in the personal knowledge graph to update the personal knowledge graph.
  • the electronic device deletes the relational knowledge whose first time interval is greater than the first threshold in the personal knowledge graph; and/or, the electronic device deletes the event knowledge whose second time interval is greater than the first threshold in the personal knowledge graph;
  • the device determines a third time interval from the third time point to the current time point according to the third time point; the electronic device deletes the entity knowledge in the personal knowledge graph whose third time interval is greater than the first threshold.
  • the electronic device 100 may add new personal data to the personal knowledge graph.
  • the electronic device 100 acquires the first personal data every fixed period, and adds the first personal data to the personal knowledge graph.
  • the electronic device 100 acquires parameter information of the first advertisement content from the advertisement server 200 .
  • the electronic device receives the parameter information of the first advertisement content from the advertisement server.
  • the parameter information may be information such as the type, link address, and size of the first advertisement content.
  • the first advertisement content may include one or more advertisements.
  • the first advertisement content is any one or more of the following: pictures, videos, texts, audios, and the like.
  • the first advertisement content may include one or more advertisements.
  • the electronic device 100 acquires parameter information of the second advertisement content from the parameter information of the first advertisement content according to the personal knowledge graph.
  • the electronic device may acquire the parameter information of the second advertisement content from the parameter information of the first advertisement content according to the personal knowledge graph in one or more of the following manners.
  • Mode 1 The electronic device retains the parameter information of all advertisements in the parameter information of the first advertisement content, and the electronic device just sorts the first advertisement content according to the predicted value of the user's preference according to the type of advertisements, and obtains the second advertisement content.
  • the parameter information of the advertisement content The parameter information of the advertisement content.
  • Manner 2 The electronic device selects the parameter information of a part of the advertisement from the parameter information of the first advertisement content to obtain the parameter information of the second advertisement content.
  • the electronic device sorts the content of the first advertisements according to the type of advertisements according to the predicted value of the user's favorite degree from high to low, and only retains the parameter information of the advertisement whose predicted value of the user's degree of likeability is higher than the first threshold, and obtains the second advertisement content.
  • the parameter information of the advertisement content is not limited to the content of the first advertisements according to the type of advertisements according to the predicted value of the user's favorite degree from high to low, and only retains the parameter information of the advertisement whose predicted value of the user's degree of likeability is higher than the first threshold, and obtains the second advertisement content.
  • the second advertisement content is any one or more of the following: pictures, videos, texts, audios, and the like.
  • the second advertisement content may include one or more advertisements.
  • the electronic device 100 trains the reordering model according to the personal knowledge graph to obtain the first model, and the electronic device obtains the parameter information of the second advertisement content from the parameter information of the first advertisement content through the first model.
  • the electronic device 100 trains a reordering model according to the personal knowledge graph, which may include the following:
  • the electronic device acquires historical user behavior and historical advertising information displayed by the electronic device.
  • the electronic device takes the historical advertisement information and the personal knowledge graph as the input of the reordering model, and the reordering model outputs the first result.
  • the electronic device compares the first result with the user's historical behavior, and modifies the parameters of the reordering model until the first result output by the reordering model is within a preset range, and the first model is obtained.
  • the electronic device 100 acquires the second advertisement content according to the parameter information of the second advertisement content.
  • the electronic device 100 acquires the second advertisement content according to the parameter information (eg, link address) of the second advertisement content.
  • the parameter information eg, link address
  • the electronic device 100 displays the second advertisement content on the display screen.
  • Manner 1 The electronic device plays one or more advertisements in the second advertisement content from high to low according to the predicted value of the user's degree of preference in the second advertisement content.
  • Manner 2 The electronic device displays the advertisement with the highest predicted value of the user's preference in the second advertisement content.
  • Mode 3 The electronic device plays one or more advertisements in the second advertisement content from high to low according to the predicted value of the user's preference in the second advertisement content, and blocks the electronic device in the second advertisement content within the first time period One or more ads played.
  • the electronic device 100 may acquire viewing data of the second advertisement content by the user; the viewing data includes the advertisement type of one or more advertisements in the second advertisement content viewed by the user and the advertisement type that the user has closed. the advertisement type of one or more advertisements in the second advertisement content;
  • the electronic device updates the first model based on the viewing data. In this way, the electronic device updates the first model according to the advertisement viewing data of the user, and the first model recommends the advertisement of the type most viewed by the user to the user when recommending advertisements to the user next time, which is more in line with the user's needs.
  • the application in the electronic device 100 can obtain the personal knowledge graph by soliciting the user's consent, and the application can perform the personalization for the user according to the personal knowledge graph. recommended.
  • the app can only obtain the personal knowledge graph by asking the user's consent, which fully respects the user's personal privacy; on the other hand, the app makes personalized recommendations for the user based on the personal knowledge graph, so that the app recommends the user for the user.
  • the content is more in line with the behavioral characteristics of users.
  • the user downloads a new application (for example, the first application) from the application store.
  • the first application will prompt the user to register personal information and log in; on the other hand, the first application The user may be prompted to select content of interest to the user, and the first application recommends the relevant content of interest to the user for viewing by the user.
  • FIG. 11A illustrates an exemplary user interface 700 on the electronic device 100 for the application menu.
  • User interface 700 may include a status bar 400, a tray 710 with frequently used application icons, a navigation bar 720, and other application icons. in:
  • a tray 710 with icons of frequently used applications may display: a phone icon 7012, a contact icon 7013, a text message icon 7014, a camera icon 7015.
  • the navigation bar 720 may include: a return button 7016 , a home screen button 7017 , a task history button 7018 and other system navigation buttons.
  • the electronic device 100 may display the previous page of the current page.
  • the electronic device 100 may display the home interface.
  • the electronic device 100 may display the tasks recently opened by the user.
  • the names of the navigation keys may also be other, which is not limited in this application. Not limited to virtual keys, each navigation key in the navigation bar 720 can also be implemented as physical keys.
  • Other application icons can be, for example: icon 7001 for clock, icon 7002 for calendar, icon 7003 for gallery, icon 7004 for memo, icon 7005 for file management, icon 7006 for email, icon 7007 for music, icon 7008 for calculator, The icon 7009 of Huawei Video, the icon 7010 of sports health, and the icon 7011 of the first application.
  • the first application is an application that the user has not used after downloading from the application store.
  • the first application may be a shopping application, an e-book application, a video application, and the like.
  • the application scenario is described with the first application being a shopping application.
  • the first application icon 7011 may receive a user's click operation, and in response to the user's click operation, the electronic device 100 displays an exemplary user interface 730 as shown in FIG. 11B .
  • User interface 730 includes status bar 400 and gender selection interface 740 .
  • the gender selection interface 740 includes controls 7101 and 7102 .
  • the control 7101 can receive a user's click operation, and in response to the user's click operation, the first application will recommend items of interest to men to the user for viewing.
  • the control 7102 can receive a user's click operation, and in response to the user's click operation, the first application will recommend items that women are interested in to the user for viewing.
  • control 7102 receives a user click operation, and in response to the user's click operation, the electronic device 100 displays an exemplary user interface 750 as shown in FIG. 11C .
  • User interface 750 includes status bar 400 and personalized recommendation selection interface 760 .
  • the personalized recommendation selection interface 760 includes a plurality of recommended topic controls and a next step control 770 .
  • the multiple recommended theme controls may include an outfit control 7501 , a sports control 7502 , a beauty control 7503 , a travel control 7504 , a food control 7505 , and a game control 7506 .
  • the multiple recommended theme controls may also include other recommended controls, which are not limited here.
  • Any one of the multiple recommended theme controls can receive a user's single-click selection, and in response to the user's single-click operation, the first application program recommends a theme that the user is interested in for viewing by the user.
  • the electronic device 100 will display an exemplary user interface of the first application, and the content displayed on the user interface of the first application is the content of beauty makeup, outfit and food selected by the user. In this way, the first application can recommend relevant content for the user according to the user's preference.
  • the first application can recommend the content that the user is interested in based on the personal knowledge graph provided in the embodiment of the present application. On the one hand, the user does not need to perform a series of operations to select the content that he is interested in. It will be more accurate to recommend content that users are interested in using a graph.
  • the first application When the user opens the first application, the first application will solicit the user's request to obtain the personal knowledge graph and other information, so that when the user authorizes the first application, the first application will obtain the personal knowledge graph and other data. User privacy is respected.
  • FIG. 12 is a schematic diagram of another system architecture according to an embodiment of the present application.
  • the system includes an electronic device 100 and an application server 300 .
  • the application server 300 sends the first content list to the first application in the electronic device 100 , the first content list may include multiple contents, and the first application receives the first content list sent by the application server 300 .
  • the electronic device 100 sends the personal knowledge graph to the first application, the first application receives the personal knowledge graph sent by the electronic device 100, and after the first application obtains the first content list, the first application
  • the knowledge graph filters multiple contents in the first content list to obtain a second content list, and the first application recommends the content in the second content list to the user for viewing.
  • FIG. 12A is an exemplary user interface 780 for the first application to obtain a user request.
  • User interface 780 includes status bar 400 and prompt box 790 .
  • the prompt box 790 displays prompt information and a disagree control 7801 and an agree control 7802.
  • the prompt information is used to prompt the user whether to agree to the first application to obtain information such as the personal knowledge graph.
  • the prompt information includes "In order to better provide related services such as browsing recommendation, publishing information, purchasing goods, etc., we will collect necessary user information (which may involve equipment, personal knowledge graph, etc.) according to the specific functional needs of your use of the service.”
  • Disagree control 7801 can accept the user's click operation. In response to the user's click operation, if the user does not agree with the first application to obtain data such as personal knowledge graphs, the first application will not recommend personalized content for the user.
  • the consent control 7802 can accept the user's click operation, and in response to the user's click operation, the user agrees to the first application to obtain data such as the personal knowledge graph, then the first application will recommend the user and the user's personal behavior based on the personal knowledge graph and other data For content with similar characteristics, for example, similar content is recommended for users to watch according to the user's preference and consumption level.
  • the first application recommends content similar to the user's personal behavioral characteristics for the user based on data such as the personal knowledge graph, and does not require the user to manually select the content that interests him.
  • the user's operation is simple; Recommending content that users are interested in based on the personal knowledge graph will be more accurate and more in line with the needs of users.
  • the term “when” may be interpreted to mean “if” or “after” or “in response to determining" or “in response to detecting" depending on the context.
  • the phrases “in determining" or “if detecting (the stated condition or event)” can be interpreted to mean “if determining" or “in response to determining" or “on detecting (the stated condition or event)” or “in response to the detection of (the stated condition or event)”.
  • the above-mentioned embodiments it may be implemented in whole or in part by software, hardware, firmware or any combination thereof.
  • software it can be implemented in whole or in part in the form of a computer program product.
  • the computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated.
  • the computer may be a general purpose computer, special purpose computer, computer network, or other programmable device.
  • the computer instructions may be stored in or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions may be downloaded from a website site, computer, server or data center Transmission to another website site, computer, server, or data center by wire (eg, coaxial cable, optical fiber, digital subscriber line) or wireless (eg, infrared, wireless, microwave, etc.).
  • the computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that includes an integration of one or more available media.
  • the usable media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media (eg, solid state drives), and the like.
  • the process can be completed by instructing the relevant hardware by a computer program, and the program can be stored in a computer-readable storage medium.
  • the program When the program is executed , which may include the processes of the foregoing method embodiments.
  • the aforementioned storage medium includes: ROM or random storage memory RAM, magnetic disk or optical disk and other mediums that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Accounting & Taxation (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Computational Linguistics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

一种广告显示方法及电子设备。该方法包括：电子设备100获取第一个人数据（S1001）；电子设备100根据第一个人数据构建个人知识图谱（S1002）；电子设备100从广告服务器200获取第一广告内容的参数信息（S1003）；电子设备100根据个人知识图谱从第一广告内容的参数信息中获取到第二广告内容的参数信息（S1004）；电子设备100根据第二广告内容的参数信息获取到所述第二广告内容（S1005）；电子设备100在显示屏上显示第二广告内容（S1006）。该方法实现了端侧和服务器侧联合的广告推荐方案。一方面，优化了广告供应商的广告投放效果，使广告供应商的广告投放更精准。另一方面，用户的个人知识图谱是在端侧构建，用户的个人知识图谱可以全方面的描述用户的行为特征，保护了用户隐私信息的安全。

Description

广告显示方法及电子设备
本申请要求于2020年09月19日提交中国专利局、申请号为202010990422.3、申请名称为“广告显示方法及电子设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及数据处理技术领域，尤其涉及广告显示方法及电子设备。
背景技术
近年来，随着网络技术的进步，互联网已成为人们生活中的重要部分。广告的内容和投放方式随着互联网的飞速发展从发生了巨大的改变。
目前，广告的投放方式已经形成了以人群为投放目标、以产品为导向的技术型投放模式。首先，服务器收集用户群体的业务数据，例如业务数据可以是用户群体观看广告的类型和时长、用户群体关闭、忽略广告的操作记录等，服务器根据用户群体的业务数据为用户群体进行群体画像，群体画像的结果可以是用户群体观看哪些类型的广告的数量最多、用户群体对哪些类型的广告不感兴趣等；服务器根据群体画像的结果对广告池中的多个广告进行筛选；服务器将排序后的广告发送至用户的电子设备。
上述广告投放方式是利用了广大用户群体的行为特征来进行广告推送，但没有考虑到个人用户的差异性，例如个人用户的喜好、需求不同，该广告推荐方式存在推荐同质化问题，无法达到最优广告投放效果。
发明内容
本申请提供了广告显示方法及电子设备。实现了端侧和服务器侧联合的广告推荐方案。一方面，优化了广告供应商的广告投放效果，使广告供应商的广告投放更精准，提高了广告供应商的经济效益。另一方面，用户的个人知识图谱是利用端侧中存储的个人数据构建的，用户的个人知识图谱可以全方面的描述用户的行为特征，而且用户的个人知识图谱在端侧建立，保护了用户隐私信息的安全。
第一方面，本申请提供了一种广告显示方法，该方法包括电子设备获取用户的第一个人数据，第一个人数据为用户的个人信息；电子设备根据第一个人数据构建个人知识图谱；个人知识图谱包括第一个人数据和第一个人数据产生的时间。之后，电子设备从广告服务器获取第一广告内容的参数信息，参数信息包括第一广告内容的类型、第一广告内容的链接地址；第一广告内容是广告服务器根据群体数据在多个广告中筛选得到的；第一广告内容包括一个或多个广告。电子设备根据个人知识图谱从第一广告内容的参数信息中获取到第二广告内容的参数信息；电子设备根据第二广告内容的参数信息获取到第二广告内容；第二广告内容包括一个或多个广告；电子设备在显示屏上显示第二广告内容。
电子设备根据个人知识图谱从第一广告内容的参数信息中获取到第二广告内容的参数信息可以采取以下方式中的一种或多种。方式一、电子设备保留第一广告内容的参数信息中的所有广告的参数信息，电子设备只是将广告的类型按照用户的喜爱度预测值从高到低将第一 广告内容进行排序，得到第二广告内容的参数信息。方式二、电子设备从第一广告内容的参数信息中筛选一部分广告的参数信息得到第二广告内容的参数信息。具体的，电子设备将广告的类型按照用户的喜爱度预测值从高到低将第一广告内容进行排序，仅保留用户的喜爱度预测值高于第一阈值的广告的参数信息，得到第二广告内容的参数信息。
在该方法中，电子设备向广告服务器发送广告推荐请求，电子设备接收到广告服务器返回的第一广告内容的参数信息。之后，该电子设备对第一广告内容的参数信息做进一步地筛选，得到第二广告内容的参数信息。
具体的，电子设备利用获取到的个人数据构建该用户的个人知识图谱，并根据该个人知识图谱训练重排序模型。当电子设备向广告服务器发送广告推荐请求之后，电子设备接收广告务服务器发送的第一广告内容的参数信息；之后，电子设备根据重排序模型对第一广告内容的参数信息做进一步地筛选，得到第二广告内容的参数信息，电子设备根据第二广告内容的参数信息获取到第二广告内容，并将第二广告内容推荐给用户观看。
该方法实现了端侧和服务器侧联合的广告推荐方案。一方面，优化了广告供应商的广告投放效果，使广告供应商的广告投放更精准，提高了广告供应商的经济效益。另一方面，用户的个人知识图谱是利用端侧中存储的个人数据构建的，用户的个人知识图谱可以全方面的描述用户的行为特征，而且用户的个人知识图谱在端侧建立，保护了用户隐私信息的安全。
结合第一方面，在第一方面的一种可能的实现方式中，电子设备根据第一个人数据构建个人知识图谱，具体包括：电子设备从第一个人数据中获取到第二个人数据；第二个人数据包括关系知识、事件知识、实体知识；电子设备将关系知识、事件知识、实体知识按照预定结构存储；电子设备根据预定结构的关系知识、预定结构的事件知识、预定结构的实体知识构建用户的个人知识图谱。这样，个人知识图谱用图形化的方式来展示个人数据之间相互联系的数据结构；并且，个人知识图谱包括第一个人数据和第一个人数据产生的时间，个人知识图谱可以表示个人数据与时间的关系，方便后续电子设备根据时间来更新个人知识图谱。
结合第一方面，在第一方面的一种可能的实现方式中，第一广告内容是一下任意一种或几种：图片、视频、文字、音频。第一广告内容还可以包括其他的内容，本申请在此不做限定。
结合第一方面，在第一方面的一种可能的实现方式中，电子每隔固定周期获取所述用户的第一个人数据。这样，电子设备可以每隔固定时间获取用户新的第一个人数据，并将新的第一个人数据加入到个人知识图谱中，从而更新个人知识图谱中的用户的个人数据。
结合第一方面，在第一方面的一种可能的实现方式中，预定结构为五元组结构；电子设备将关系知识按照预定结构存储，具体包括：电子设备将关系知识按照第一五元组结构进行存储；第一五元组结构为“第一实体-关系-第二实体-第一时间点-第一时间区间”；关系表征第一实体与第二实体的关系，第一时间点为第一实体与述第二实体建立关系的时间，第一时间区间为第一时间点到当前时间点的间隔时间。这样，电子设备将用户的关系知识存储为预定结构，方便后续构建个人知识图谱。并且，表示关系知识的第一五元组包括第一时间点和第一时间区间，电子设备可以根据第一时间点和第一时间区间来更新用户的关系知识。
结合第一方面，在第一方面的一种可能的实现方式中，预定结构为五元组结构；电子设备将事件知识按照预定结构存储，具体包括：电子设备将事件知识按照第二五元组结构进行存储；第二五元组结构为“事件-论元-逻辑关系-第二时间点-第二时间区间”；论元为事件的发生动作，逻辑关系表征事件与论元的关系，第二时间点为事件发生的时间，第二时间区间为第二时间点到当前时间点的间隔时间。这样，电子设备将用户的事件知识存储为预定结构， 方便后续构建个人知识图谱。并且，表示事件知识的第二五元组包括第二时间点和第二时间区间，电子设备可以根据第二时间点和第二时间区间来更新用户的事件知识。
结合第一方面，在第一方面的一种可能的实现方式中，预定结构为五元组结构；电子设备将实体知识按照预定结构存储，具体包括：电子设备将实体知识按照第三五元组结构进行存储；第三五元组结构为“第三实体：第三时间点-第一关联权重-第四实体-第二关联权重-第五实体”；第三时间点为第三实体的发生时间，第一关联权重为第三实体与第四实体的关联程度，第二关联权重为第四实体与第五实体的关联程度。这样，电子设备将用户的实体知识存储为预定结构，方便后续构建个人知识图谱。并且，表示实体知识的第三五元组包括第三时间点和第三时间区间，电子设备可以根据第三时间点和第三时间区间来更新用户的实体知识。
结合第一方面，在第一方面的一种可能的实现方式中，电子设备删除个人知识图谱中第一时间区间大于第一阈值的关系知识；和/或，电子设备删除个人知识图谱中第二时间区间大于第一阈值的事件知识；和/或，电子设备根据第三时间点确定出第三时间点到当前时间点的第三时间区间；电子设备删除个人知识图谱中第三时间区间大于第一阈值的实体知识。这样，电子设备可以根据时间来删除个人知识图谱中时间区间大于第一阈值的用户知识，去除时间久远的用户知识，个人知识图谱更能表征用户最近一段时间的行为特征。
结合第一方面，在第一方面的一种可能的实现方式中，在电子设备根据用户的第一个人数据构建个人知识图谱之后，方法还包括：电子设备获取到用户历史行为和电子设备显示的历史广告信息；电子设备将历史广告信息、个人知识图谱作为重排序模型的输入，重排序模型输出第一结果；电子设备将第一结果与用户历史行为比较，并修改重排序模型的参数，直至重排序模型输出的第一结果在预设范围内，得到第一模型；电子设备根据个人知识图谱从第一广告内容的参数信息中获取到第二广告内容的参数信息，具体包括：电子设备根据第一模型从第一广告内容的参数信息中获取到第二广告内容的参数信息。这样，电子设备根据个人知识图谱训练重排序模型，得到第一模型。电子设备可以根据第一模型从第一广告内容的参数信息中获取到第二广告内容的参数信息。
结合第一方面，在第一方面的一种可能的实现方式中，电子设备根据第一模型从第一广告内容的参数信息中获取到第二广告内容的参数信息，具体包括：电子设备根据第一模型将第一广告内容的类型按照用户的喜爱度预测值从高到低进行排序，得到第二广告内容的参数信息；或者，电子设备根据第一模型将第一广告内容的类型按照用户的喜爱度预测值从高到低进行排序，并获取用户的喜爱度预测值高于第一阈值的广告的类型，得到第二广告内容的参数信息。这样，电子设备按照用户的喜爱度预测值从第一广告内容的参数信息中获取到第二广告内容的参数信息，使电子设备显示的广告更符合用户的喜好。这样，可以提高广告推荐效果。
结合第一方面，在第一方面的一种可能的实现方式中，电子设备从第一个人数据中获取到第二个人数据之前，方法还包括：电子设备将第一个人数据转化为文本信息；电子设备对文本信息进行断句、分词和词性标注；电子设备从第一个人数据中获取到第二个人数据，具体包括：电子设备获取文本信息中属于预设词性的词语。这样，电子设备去除第一个人数据中不能表征用户行为特征的数据。电子设备去除无用的数据，得到的第二个人数据更能刻画用户的行为特征，使得构建的个人知识图谱更能准确的表征用户的行为特征。
结合第一方面，在第一方面的一种可能的实现方式中，电子设备获取文本信息中属于预设词性的词语之后，方法还包括：电子设备获取文本信息中出现次数为一次的词语；若文本 信息中有出现两个及两个以上相同的词语，电子设备获取文本信息中有出现两个及两个以上相同的词语中的一个词语，得到第二个人数据。这样，电子设备去除重复的数据，减少了数据的冗余。
结合第一方面，在第一方面的一种可能的实现方式中，用户的个人信息包括以下一项或多项：性别、年龄、性格、爱好、人际关系、收入、通讯录信息、通话记录、短信、备忘录信息、居住的地址、居住的地址的天气情况。
结合第一方面，在第一方面的一种可能的实现方式中，电子设备在显示屏的广告显示区域显示第二广告内容，具体包括：电子设备按照第二广告内容中用户的喜爱度预测值从高到低播放第二广告内容中的一个或多个广告；或者，电子设备播放第二广告内容中所用户的喜爱度预测值最高的广告；或者，电子设备按照第二广告内容中用户的喜爱度预测值从高到低播放投放广告中的一个或多个广告，并屏蔽掉第二广告内容中电子设备在第一时间段内播放过的一个或多个广告。这样，电子设备按照用户的喜爱度预测值从高到低播放第二广告内容中的一个或多个广告或播放用户的喜爱度预测值最高的广告，更符合用户的喜好，那么用户观看广告的可能性越高。并且，电子设备屏蔽掉电子设备在第一时间段内播放过的一个或多个广告，避免在短时间内重复推荐相同的广告，影响用户体验。
结合第一方面，在第一方面的一种可能的实现方式中，电子设备在显示屏的广告显示区域显示第二广告内容之后，方法还包括：电子设备获取用户对第二广告内容的观看数据；观看数据包括用户观看了第二广告内容中的一个或多个广告的广告类型和用户关闭了第二广告内容中的一个或多个广告的广告类型；电子设备根据观看数据更新第一模型。这样，电子设备根据用户的观看广告的数据来更新第一模型，第一模型会在下一次给用户推荐广告时将用户观看次数最多的类型的广告推荐给用户，这样，更符合用户的需求。
第二方面，本申请提供了一种电子设备，电子设备包括一个或多个处理器、一个或多个存储器、显示屏；一个或多个存储器、显示屏与一个或多个处理器耦合，一个或多个存储器用于存储计算机程序代码，计算机程序代码包括计算机指令，一个或多个处理器调用计算机指令以使得电子设备执行上述第一方面以及结合上述第一方面中的任意一种实现方式所提供的广告显示方法。
第三方面，本申请提供了一种计算机存储介质，计算机可读存储介质存储有计算机程序，当该计算机程序被处理器执行时，处理器执行上述第一方面以及结合上述第一方面中的任意一种实现方式所提供的广告显示方法。
第四方面，本申请实施例提供了一种计算机程序产品，计算机可读存储介质存储有计算机程序，当该计算机程序被处理器执行时，处理器执行上述第一方面以及结合上述第一方面中的任意一种实现方式所提供的广告显示方法。
在该方法中，电子设备利用获取到的个人数据构建该用户的个人知识图谱，并根据该个人知识图谱训练重排序模型。当电子向广告服务器发送广告推荐请求之后，电子设备接收广告务服务器发送的第一广告内容的参数信息；之后，电子设备根据重排序模型对第一广告内容的参数信息做进一步地筛选，得到第二广告内容的参数信息，电子设备根据第二广告内容的参数信息获取到第二广告内容，并将第二广告内容推荐给用户观看。
该方法实现了端侧和服务器侧联合的广告推荐方案。一方面，优化了广告供应商的广告投放效果，使广告供应商的广告投放更精准，提高了广告供应商的经济效益。另一方面，用户的个人知识图谱是利用端侧中存储的个人数据构建的，用户的个人知识图谱可以全方面的描述用户的行为特征，而且用户的个人知识图谱在端侧建立，保护了用户隐私信息的安全。
附图说明
图1为本申请实施例提供的一种广告推荐系统示意图；
图2为本申请实施例提供的一种电子设备100的结构示意图；
图3为本申请实施例提供的一种电子设备100的软件结构框图；
图4为本申请实施例提供的一种广告服务器200的硬件结构示意图；
图5为本申请实施例提供的另一种广告推荐系统架构示意图；
图6A-图6E为本申请实施例提供的一组应用程序界面图；
图7为本申请实施例提供的一种广告显示方法流程示意图；
图8为本申请实施例提供的一种以图的形式表示构建的个人知识图谱的示意图；
图9A-图9C为本申请实施例提供的一组UI图；
图10为本申请实施例提供的另一种广告显示方法流程示意图；
图11A-图11C为本申请实施例提供的另一组UI图；
图12为本申请实施例提供的另一种系统架构示意图；
图12A为本申请实施例提供的一种UI图。
具体实施方式
下面将结合附图对本申请实施例中的技术方案进行清除、详尽地描述。其中，在本申请实施例的描述中，除非另有说明，“/”表示或的意思，例如，A/B可以表示A或B；文本中的“或”仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况，另外，在本申请实施例的描述中，“多个”是指两个或多于两个。
以下，术语“第一”、“第二”仅用于描述目的，而不能理解为暗示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括一个或者更多个该特征，在本申请实施例的描述中，除非另有说明，“多个”的含义是两个或两个以上。
本申请的说明书和权利要求书及附图中的术语“用户界面(user interface，UI)”，是应用程序或操作系统与用户之间进行交互和信息交换的介质接口，它实现信息的内部形式与用户可以接受形式之间的转换。应用程序的用户界面是通过java、可扩展标记语言(extensible markup language，XML)等特定计算机语言编写的源代码，界面源代码在终端设备上经过解析，渲染，最终呈现为用户可以识别的内容，比如图片、文字、按钮等控件。控件(control)也称为部件(widget)，是用户界面的基本元素，典型的控件有工具栏(toolbar)、菜单栏(menu bar)、文本框(text box)、按钮(button)、滚动条(scrollbar)、图片和文本。界面中的控件的属性和内容是通过标签或者节点来定义的，比如XML通过<Textview>、<ImgView>、<VideoView>等节点来规定界面所包含的控件。一个节点对应界面中一个控件或属性，节点经过解析和渲染之后呈现为用户可视的内容。此外，很多应用程序，比如混合应 用(hybrid application)的界面中通常还包含有网页。网页，也称为页面，可以理解为内嵌在应用程序界面中的一个特殊的控件，网页是通过特定计算机语言编写的源代码，例如超文本标记语言(hyper text markup language，HTML)，层叠样式表(cascading style sheets，CSS)，java脚本(JavaScript，JS)等，网页源代码可以由浏览器或与浏览器功能类似的网页显示组件加载和显示为用户可识别的内容。网页所包含的具体内容也是通过网页源代码中的标签或者节点来定义的，比如HTML通过<p>、<img>、<video>、<canvas>来定义网页的元素和属性。
用户界面常用的表现形式是图形用户界面(graphic user interface，GUI)，是指采用图形方式显示的与计算机操作相关的用户界面。它可以是在电子设备的显示屏中显示的一个图标、窗口、控件等界面元素，其中控件可以包括图标、按钮、菜单、选项卡、文本框、对话框、状态栏、导航栏、Widget等可视的界面元素。
为了便于理解本申请，下面对本申请涉及的术语进行解释。
文件系统：文件系统用于存储电子设备中各个应用程序在运行过程当中产生的非结构化的个人数据。非结构化的个人数据就是不能用二维逻辑表来表现的数据。非结构化的个人数据可以是例如文档、图片、视频、文本等数据。示例性的，非结构化的个人数据可以是相机应用在运行过程中产生的数据。相机应用拍摄的图片和视频存储在文件系统里面，那么相机应用拍摄的图片和视频就是非结构化的个人数据。
数据服务：数据服务用于存储电子设备中各个应用程序在运行过程当中产生的结构化的个人数据。结构化的个人数据就是可以用统一的结构来表示的数据。示例性的，结构化的个人数据可以是上述通讯录应用运行过程中产生的数据。例如，通讯录应用中存储的用户联系人名字以及用户联系人电话，在数据服务中，用户联系人名字与用户联系人电话是一一对应保存的。用户联系人名字以及用户联系人电话属于结构化的个人数据。
个人数据：个人数据包括涉及个人隐私的数据。
具体的，个人数据可以是电子设备运行各个应用程序的过程当中产生的涉及个人隐私的数据，各个应用程序的过程当中产生的数据保存至文件系统和/或数据服务中了。个人数据还可以是电子设备在获得用户的授权后，直接从各个应用程序中获取的涉及个人隐私的数据。例如通讯类应用、短信应用、通讯录应用、备忘录应用、天气应用、购物应用等。
一方面，电子设备中的各个应用程序在运行过程中产生的数据存放至数据服务和/或文件系统中，电子设备可以从数据服务和/或文件系统中获取到用户的个人数据。
另一方面，电子设备中的应用程序可以获得用户的授权，在获得用户授权之后，电子设备可以从各个应用程序获取到用户的个人数据。需要说明的是，电子设备直接从应用程序中获取用户的个人数据也可以分为结构化的个人数据和非结构化的个人数据。
群体数据：群体数据包括用户群体中的多个用户的不涉及用户隐私的数据，例如用户观看广告时产生的业务数据。例如可包括以下一项或多项：用户经常点击的广告、用户从未点击过的广告、用户观看广告的时长、用户关闭的广告等等。
个人知识图谱：个人知识图谱是根据用户的个人数据构建的、是用图形化的方式来展示 个人数据之间相互联系的数据结构。
由于不同用户的个人数据是不同的，不同用户的个人知识图谱是不一样的。
目前，群体知识图谱都是根据用户群体的群体数据构建的，即知识图谱是表示用户群体的群体数据之间的相互联系的数据结构。群体知识图谱不能表示个人用户的行为特征。
本申请中，电子设备可以根据每个用户的个人数据来构建该用户的个人知识图谱。个人知识图谱的具体构建过程可参考后续方法实施例的详细描述，暂不赘述。
群体画像：群体画像是利用群体数据生成的针对用户群体的标签。
用户群体的标签可以包括但不限于：该用户群体喜欢浏览的广告的类型、该用户群体忽略广告最多的广告类型、用户群体关闭广告最多的广告类型、用户群体举报广告最多的广告类型等等。
在一些实施例中，用户群体是不分性别、不分年龄、不分地域的所有用户的集合。
在一些实施例中，用户群体可以根据性别分类，例如用户群体可以分为女性用户群体和男性用户群体。或者用户群体也可以分为各个年龄段的用户群体等。
示例性的，当用户群体分为女性用户群体和男性用户群体时，群体画像可以分为女性用户群体画像和男性用户群体画像。具体的，群体数据可以分为男性用户群体的数据和女性用户群体的数据。根据男性用户群体的数据为男性用户群体进行画像，即对男性用户群体贴标签，例如，男性用户群体对车类广告最感兴趣、对美妆类广告不感兴趣等。根据女性用户群体的数据为女性用户群体进行画像，即对女性用户群体贴标签，例如，女性用户群体对服饰类广告和美妆类广告感兴趣、对运动类广告不感兴趣等。
示例性的，当用户群体分为各个年龄段的用户群体时，群体画像还可以根据各个年龄段的用户群体对用户群体画像。具体的，群体数据可以按照年龄段进行分类。例如将用户年龄在0-20岁之间的用户群体数据分为一类，将用户年龄在21-35岁之间的用户群体数据分为一类，将用户年龄在36-50岁之间的用户群体数据分为一类，将用户年龄在51-70岁之间的用户群体数据分为一类。根据各个年龄段的用户群体数据对各个年龄段的用户群体分别进行画像，即为各个年龄段的用户群体贴标签。例如，年龄在0-20岁之间的用户群体对玩具类广告最感兴趣、年龄在21-35岁之间的用户群体对电子产品类广告最感兴趣、年龄在36-50岁之间的用户群体对护肤护发类广告最感兴趣、年龄在51-70岁之间的用户群体对健康保健类广告最感兴趣等。
广告：广告是一种向大众传播信息的手段。广告可以分为公益性广告和盈利性广告。公益广告是不以营利为目的而为社会提供免费服务的广告活动。盈利性广告可以包括各种应用、以及产品或者一些品牌的推广。例如盈利性广告可以是美妆、美食，类广告、音乐试听推荐广告、视频推荐广告、小说推荐广告、电影推荐广告、应用程序下载推荐广告等等。
目前，广告投放流程可以包括以下步骤：
首先，广告服务器接收应用程序服务器发送的广告推荐请求，响应于广告推荐请求，广告服务器发送获取广告请求至厂商服务器(例如手机厂商服务器)。
其中，广告服务器的作用是获取各个厂商的广告，并对广告进行筛选，之后，广告服务器将筛选得到的广告发送到各个应用程序进行显示。
厂商服务器可以有多个，例如手机厂商服务器、车厂商服务器、美妆品牌厂商服务器等，多个厂商服务器接收广告服务器发送的获取广告请求，当多个厂商服务器中的任意一个厂商 服务器有待发广告时、任意一个厂商服务器可以将待发的广告发送至广告服务器。
广告服务器接收任意一个厂商服务器发送的广告，广告服务器对多个广告进行筛选(例如按照广告的价格从高到低进行筛选)，得到广告集。
在一些实施例中，任意一个厂商服务器发送给广告服务器的广告可能出现重复，广告服务器还可以对获取的广告进行去重，避免出现重复的广告。
广告服务器根据用户群体的群体数据对广告集中的广告做进一步的筛选，得到广告列表，广告服务器将广告列表发送至应用程序，应用程序显示广告列表中的广告。
粗排广告：粗排广告是一个或多个广告的集合，粗排广告是广告服务器根据群体数据按照用户群体对广告的感兴趣程度在大量的广告中筛选得到的。
投放广告：投放广告是一个或多个广告的集合，是电子设备根据用户的个人数据按个人可能点击广告的概率在粗排广告中筛选得到的。其中，投放广告的数量可以与粗排广告的数量相同。投放广告的数量也可以比粗排广告的数量少，因为电子设备可以根据用户的个人数据将粗排广告中的某些广告滤除掉。
为了便于理解本申请，下面对广告推荐系统进行介绍。
如图1所示，图1为广告推荐系统示意图。系统10可以包括图1中所示的电子设备100、广告服务器200和应用程序服务器300。
其中，电子设备100可用于检测用户操作，并响应于该用户操作，电子设备100向应用程序服务器300发送用户请求。
其中，该用户操作例如可以是：电子设备100检测到用户开启了应用程序，或者电子设备100检测到用户下一个视频了应用程序的当前浏览页面(例如电子设备100检测到用户单指向下滑动来下一个视频当前浏览界面)，则电子设备100发送用户请求至应用程序服务器300。
应用程序服务器300可用于接收并响应于用户请求，应用程序服务器300发送广告推荐请求至广告服务器200。
广告服务器200可用于根据用户的群体画像对广告集中的广告进行筛选，得到粗排广告。
广告服务器200，还可用于将粗排广告发送至电子设备100。
电子设备100，还可用于接收广告服务器200发送的粗排广告，并对粗排广告做进一步的筛选，得到投放广告，之后，电子设备100将投放广告推荐给用户观看。
需要说明的是，图1中的广告推荐系统架构只是本申请实施例中的一种示例性的实施方式，本申请实施例中的广告推荐系统架构包括但不仅限于以上广告推荐系统架。
目前，广告投放一般有两种方式。方式一：应用程序服务器会收集多个用户观看广告的业务数据，并将用户群体的业务数据上传至广告服务器。广告服务器根据用户群体的业务数据多个广告进行筛选，得到用户群体感兴趣的广告列表。广告服务器将用户群体感兴趣的广告发送至用户。这种广告投放方式是利用了用户群体的业务数据来进行广告，这种广告推送方式没有考虑到个人用户的差异性，存在推荐同质化问题。方式二、应用服务器会收集个人用户的搜索信息和浏览信息，应用服务器提取出搜索信息和浏览信息中的关键词对个人用户 进行内容推荐。例如，个人用户在购物应用程序上搜索了一个想要购买的物品(例如耳机)，在个人用户下次使用该购物应用程序时，购物应用程序会向个人用户推荐多种耳机物品信息。这种推荐方式会收集用户的搜索信息和浏览信息，对用户来说，会导致用户的个人隐私泄露。
为了解决上述问题，本申请以下实施例提供的了一种广告显示方法。在该方法中，电子设备可以接收用户操作，响应于用户操作，电子设备向应用程序服务器发送用户请求，应用程序服务器向广告服务器发送广告推荐请求，电子设备接收到广告服务器返回的粗排广告(第一广告内容)。之后，该电子设备对粗排广告做进一步地筛选，得到投放广告(第二广告内容)。粗排广告、投放广告的数量均可以为一个或多个。
在一些实施例中，也可以是电子设备直接接收到广告服务器返回的粗排广告(第一广告内容)。之后，该电子设备对粗排广告做进一步地筛选，得到投放广告(第二广告内容)。电子设备不必向应用程序服务器发送用户请求。本身请在此不做限定。
具体的，电子设备利用获取到的个人数据构建该用户的个人知识图谱，并根据该个人知识图谱训练重排序模型。当应用程序服务器向广告服务器发送广告推荐请求之后，电子设备接收广告务服务器发送的粗排广告；之后，电子设备根据重排序模型对粗排广告做进一步地筛选，得到投放广告，电子设备将投放广告推荐给用户观看。
在一些实施例中，电子设备接收到广告服务器的可以是第一广告内容的参数信息。参数信息可以是第一广告内容的类型、链接地址、大小等信息。电子设备根据个人知识图谱从第一广告内容的参数信息中得到第二广告内容的参数信息。电子设备根据第二广告内容的参数信息从广告服务器获取到第二广告内容。本申请在此不做限定。
该方法实现了端侧和服务器侧联合的广告推荐方案。一方面，优化了广告供应商的广告投放效果，使广告供应商的广告投放更精准，提高了广告供应商的经济效益。另一方面，用户的个人知识图谱是利用端侧中存储的个人数据构建的，用户的个人知识图谱可以全方面的描述用户的行为特征，而且用户的个人知识图谱在端侧建立，保护了用户隐私信息的安全。
接下来，介绍本申请实施例的提及的电子设备的硬件架构。
图2示出了电子设备100的结构示意图。
下面以电子设备100为例对实施例进行具体说明。电子设备100的设备类型可以包括手机、电视、平板电脑、音箱、手表、桌面型计算机、膝上计算机、手持计算机、笔记本电脑、超级移动个人计算机(ultra-mobilepersonalcomputer，UMPC)、上网本，以及个人数字助理(personal digitalassistant，PDA)、增强现实(augmentedreality，AR)/虚拟现实(virtualreality，VR)设备等。本申请实施例对电子设备100的设备类型不做特殊限制。
应该理解的是，图2所示电子设备100仅是一个范例，并且电子设备100可以具有比图2中所示的更多的或者更少的部件，可以组合两个或多个的部件，或者可以具有不同的部件配置。图中所示出的各种部件可以在包括一个或多个信号处理和/或专用集成电路在内的硬件、软件、或硬件和软件的组合中实现。
电子设备100可以包括：处理器110，外部存储器接口120，内部存储器121，通用串行总线(universal serial bus，USB)接口130，充电管理模块140，电源管理模块141，电池142，天线1，天线2，移动通信模块150，无线通信模块160，音频模块170，扬声器170A，受话器170B，麦克风170C，耳机接口170D，传感器模块180，按键190，马达191，指示器192，摄像头193，显示屏194，以及用户标识模块(subscriber identificationmodule，SIM)卡接口195等。其中传感器模块180可以包括压力传感器180A，陀螺仪传感器180B，气压传 感器180C，磁传感器180D，加速度传感器180E，距离传感器180F，接近光传感器180G，指纹传感器180H，温度传感器180J，触摸传感器180K，环境光传感器180L，骨传导传感器180M等。
可以理解的是，本发明实施例示意的结构并不构成对电子设备100的具体限定。在本申请另一些实施例中，电子设备100可以包括比图示更多或更少的部件，或者组合某些部件，或者拆分某些部件，或者不同的部件布置。图示的部件可以以硬件，软件或软件和硬件的组合实现。
处理器110可以包括一个或多个处理单元，例如：处理器110可以包括应用处理器(application processor，AP)，调制解调处理器，图形处理器(graphics processing unit，GPU)，图像信号处理器(image signal processor，ISP)，控制器，存储器，视频编解码器，数字信号处理器(digital signal processor，DSP)，基带处理器，或神经网络处理器(neural-network processing unit，NPU)等。其中，不同的处理单元可以是独立的器件，也可以集成在一个或多个处理器中。
其中，控制器可以是电子设备100的神经中枢和指挥中心。控制器可以根据指令操作码和时序信号，产生操作控制信号，完成取指令和执行指令的控制。
处理器110中还可以设置存储器，用于存储指令和数据。在一些实施例中，处理器110中的存储器为高速缓冲存储器。该存储器可以保存处理器110刚用过或循环使用的指令或数据。如果处理器110需要再次使用该指令或数据，可从存储器中直接调用。避免了重复存取，减少了处理器110的等待时间，因而提高了系统的效率。
在一些实施例中，处理器110可以包括一个或多个接口。接口可以包括集成电路(inter-integrated circuit，I2C)接口，集成电路内置音频(inter-integrated circuit sound，I2S)接口，脉冲编码调制(pulse code modulation，PCM)接口，通用异步收发传输器(universal asynchronous receiver/transmitter，UART)接口，移动产业处理器接口(mobile industry processor interface，MIPI)，通用输入输出(general-purpose input/output，GPIO)接口，用户标识模块(subscriber identity module，SIM)接口，或通用串行总线(universal serial bus，USB)接口等。
充电管理模块140用于从充电器接收充电输入。其中，充电器可以是无线充电器，也可以是有线充电器。在一些有线充电的实施例中，充电管理模块140可以通过USB接口130接收有线充电器的充电输入。在一些无线充电的实施例中，充电管理模块140可以通过电子设备100的无线充电线圈接收无线充电输入。充电管理模块140为电池142充电的同时，还可以通过电源管理模块141为电子设备供电。
电源管理模块141用于连接电池142，充电管理模块140与处理器110。电源管理模块141接收电池142和/或充电管理模块140的输入，为处理器110，内部存储器121，外部存储器，显示屏194，摄像头193，和无线通信模块160等供电。电源管理模块141还可以用于监测电池容量，电池循环次数，电池健康状态(漏电，阻抗)等参数。在其他一些实施例中，电源管理模块141也可以设置于处理器110中。在另一些实施例中，电源管理模块141和充电管理模块140也可以设置于同一个器件中。
电子设备100的无线通信功能可以通过天线1，天线2，移动通信模块150，无线通信模块160，调制解调处理器以及基带处理器等实现。
天线1和天线2用于发射和接收电磁波信号。电子设备100中的每个天线可用于覆盖单个或多个通信频带。不同的天线还可以复用，以提高天线的利用率。
移动通信模块150可以提供应用在电子设备100上的包括2G/3G/4G/5G等无线通信的解决方案。移动通信模块150可以包括至少一个滤波器，开关，功率放大器，低噪声放大器(low noiseamplifier，LNA)等。移动通信模块150可以由天线1接收电磁波，并对接收的电磁波进行滤波，放大等处理，传送至调制解调处理器进行解调。移动通信模块150还可以对经调制解调处理器调制后的信号放大，经天线1转为电磁波辐射出去。
调制解调处理器可以包括调制器和解调器。其中，调制器用于将待发送的低频基带信号调制成中高频信号。解调器用于将接收的电磁波信号解调为低频基带信号。
无线通信模块160可以提供应用在电子设备100上的包括无线局域网(wireless local area networks，WLAN)(如无线保真(wireless fidelity，Wi-Fi)网络)，蓝牙(bluetooth，BT)，全球导航卫星系统(global navigation satellite system，GNSS)，调频(frequency modulation，FM)，近距离无线通信技术(near field communication，NFC)，红外技术(infrared，IR)等无线通信的解决方案。无线通信模块160可以是集成至少一个通信处理模块的一个或多个器件。无线通信模块160经由天线2接收电磁波，将电磁波信号调频以及滤波处理，将处理后的信号发送到处理器110。无线通信模块160还可以从处理器110接收待发送的信号，对其进行调频，放大，经天线2转为电磁波辐射出去。
电子设备100通过GPU，显示屏194，以及应用处理器等实现显示功能。显示屏194用于显示图像，视频等。显示屏194包括显示面板。
电子设备100可以通过ISP，摄像头193，视频编解码器，GPU，显示屏194以及应用处理器等实现拍摄功能。
在一些实施例中，移动通信模块150提供的无线通信的解决方案可使得电子设备可以与网络中的设备(如广告服务器)通信，无线通信模块160提供的WLAN无线通信的解决方案也可使得电子设备可以与网络中的设备(如广告服务器)通信。
在一些实施例中，电子设备100可通过无线通信模块160发送广告推荐请求至与广告服务器建立通信连接，电子设备100还可通过无线通信模块160接收广告服务器发送的粗排广告，电子设备100还可通过处理器110对粗排广告进行筛选，得到投放广告，电子设备100还可用于通过显示屏194显示投放广告给用户观看。
ISP用于处理摄像头193反馈的数据。例如，拍照时，打开快门，光线通过镜头被传递到摄像头感光元件上，光信号转换为电信号，摄像头感光元件将电信号传递给ISP处理，转化为肉眼可见的图像。
摄像头193用于捕获静态图像或视频。物体通过镜头生成光学图像投射到感光元件。
数字信号处理器用于处理数字信号，除了可以处理数字图像信号，还可以处理其他数字信号。
NPU为神经网络(neural-network，NN)计算处理器，通过借鉴生物神经网络结构，例如借鉴人脑神经元之间传递模式，对输入信息快速处理，还可以不断的自学习。
外部存储器接口120可以用于连接外部存储卡，例如Micro SD卡，实现扩展电子设备100的存储能力。
内部存储器121可以用于存储计算机可执行程序代码，可执行程序代码包括指令。处理器110通过运行存储在内部存储器121的指令，从而执行电子设备100的各种功能应用以及数据处理。
电子设备100可以通过音频模块170，扬声器170A，受话器170B，麦克风170C，耳机接口170D，以及应用处理器等实现音频功能。例如音乐播放，录音等。
音频模块170用于将数字音频信息转换成模拟音频信号输出，也用于将模拟音频输入转换为数字音频信号。音频模块170还可以用于对音频信号编码和解码。
扬声器170A，也称“喇叭”，用于将音频电信号转换为声音信号。电子设备100可以通过扬声器170A收听音乐，或收听免提通话。
受话器170B，也称“听筒”，用于将音频电信号转换成声音信号。当电子设备100接听电话或语音信息时，可以通过将受话器170B靠近人耳接听语音。
麦克风170C，也称“话筒”，“传声器”，用于将声音信号转换为电信号。当拨打电话或发送语音信息时，用户可以通过人嘴靠近麦克风170C发声，将声音信号输入到麦克风170C。电子设备100可以设置至少一个麦克风170C。在另一些实施例中，电子设备100可以设置两个麦克风170C，除了采集声音信号，还可以实现降噪功能。在另一些实施例中，电子设备100还可以设置三个，四个或更多麦克风170C，实现采集声音信号，降噪，还可以识别声音来源，实现定向录音功能等。
本实施例中电子设备100通过麦克风170C采集声音信号，并将声音信号传送至电子设备100中的应用程序中。
耳机接口170D用于连接有线耳机。耳机接口170D可以是USB接口130，也可以是3.5mm的开放移动电子设备平台(open mobile terminal platform，OMTP)标准接口，美国蜂窝电信工业协会(cellular telecommunications industry association of the USA，CTIA)标准接口。
压力传感器180A用于感受压力信号，可以将压力信号转换成电信号。在一些实施例中，压力传感器180A可以设置于显示屏194。压力传感器180A的种类很多，如电阻式压力传感器，电感式压力传感器，电容式压力传感器等。电容式压力传感器可以是包括至少两个具有导电材料的平行板。当有力作用于压力传感器180A，电极之间的电容改变。电子设备100根据电容的变化确定压力的强度。当有触摸操作作用于显示屏194，电子设备100根据压力传感器180A检测所述触摸操作强度。电子设备100也可以根据压力传感器180A的检测信号计算触摸的位置。在一些实施例中，作用于相同触摸位置，但不同触摸操作强度的触摸操作，可以对应不同的操作指令。例如：当有触摸操作强度小于第一压力阈值的触摸操作作用于短消息应用图标时，执行查看短消息的指令。当有触摸操作强度大于或等于第一压力阈值的触摸操作作用于短消息应用图标时，执行新建短消息的指令。
陀螺仪传感器180B可以用于确定电子设备100的运动姿态。在一些实施例中，可以通过陀螺仪传感器180B确定电子设备100围绕三个轴(即，x，y和z轴)的角速度。陀螺仪传感器180B可以用于拍摄防抖。示例性的，当按下快门，陀螺仪传感器180B检测电子设备100抖动的角度，根据角度计算出镜头模组需要补偿的距离，让镜头通过反向运动抵消电子设备100的抖动，实现防抖。陀螺仪传感器180B还可以用于导航，体感游戏场景。
气压传感器180C用于测量气压。在一些实施例中，电子设备100通过气压传感器180C测得的气压值计算海拔高度，辅助定位和导航。
磁传感器180D包括霍尔传感器。电子设备100可以利用磁传感器180D检测翻盖皮套的开合。在一些实施例中，当电子设备100是翻盖机时，电子设备100可以根据磁传感器180D检测翻盖的开合。进而根据检测到的皮套的开合状态或翻盖的开合状态，设置翻盖自动解锁等特性。
加速度传感器180E可检测电子设备100在各个方向上(一般为三轴)加速度的大小。当电子设备100静止时可检测出重力的大小及方向。还可以用于识别电子设备姿态，应用于横竖 屏切换，计步器等应用。
距离传感器180F，用于测量距离。电子设备100可以通过红外或激光测量距离。在一些实施例中，拍摄场景，电子设备100可以利用距离传感器180F测距以实现快速对焦。
接近光传感器180G可以包括例如发光二极管(LED)和光检测器，例如光电二极管。发光二极管可以是红外发光二极管。电子设备100通过发光二极管向外发射红外光。电子设备100使用光电二极管检测来自附近物体的红外反射光。当检测到充分的反射光时，可以确定电子设备100附近有物体。当检测到不充分的反射光时，电子设备100可以确定电子设备100附近没有物体。电子设备100可以利用接近光传感器180G检测用户手持电子设备100贴近耳朵通话，以便自动熄灭屏幕达到省电的目的。接近光传感器180G也可用于皮套模式，口袋模式自动解锁与锁屏。
环境光传感器180L用于感知环境光亮度。电子设备100可以根据感知的环境光亮度自适应调节显示屏194亮度。环境光传感器180L也可用于拍照时自动调节白平衡。环境光传感器180L还可以与接近光传感器180G配合，检测电子设备100是否在口袋里，以防误触。
指纹传感器180H用于采集指纹。电子设备100可以利用采集的指纹特性实现指纹解锁，访问应用锁，指纹拍照，指纹接听来电等。
温度传感器180J用于检测温度。在一些实施例中，电子设备100利用温度传感器180J检测的温度，执行温度处理策略。例如，当温度传感器180J上报的温度超过阈值，电子设备100执行降低位于温度传感器180J附近的处理器的性能，以便降低功耗实施热保护。在另一些实施例中，当温度低于另一阈值时，电子设备100对电池142加热，以避免低温导致电子设备100异常关机。在其他一些实施例中，当温度低于又一阈值时，电子设备100对电池142的输出电压执行升压，以避免低温导致的异常关机。
触摸传感器180K，也可称触控面板或触敏表面。触摸传感器180K可以设置于显示屏194，由触摸传感器180K与显示屏194组成触摸屏，也称“触控屏”。触摸传感器180K用于检测作用于其上或附近的触摸操作。触摸传感器可以将检测到的触摸操作传递给应用处理器，以确定触摸事件类型。可以通过显示屏194提供与触摸操作相关的视觉输出。在另一些实施例中，触摸传感器180K也可以设置于电子设备100的表面，与显示屏194所处的位置不同。
骨传导传感器180M可以获取振动信号。在一些实施例中，骨传导传感器180M可以获取人体声部振动骨块的振动信号。骨传导传感器180M也可以接触人体脉搏，接收血压跳动信号。在一些实施例中，骨传导传感器180M也可以设置于耳机中，结合成骨传导耳机。音频模块170可以基于所述骨传导传感器180M获取的声部振动骨块的振动信号，解析出语音信号，实现语音功能。应用处理器可以基于所述骨传导传感器180M获取的血压跳动信号解析心率信息，实现心率检测功能。
按键190包括开机键，音量键等。按键190可以是机械按键。也可以是触摸式按键。电子设备100可以接收按键输入，产生与电子设备100的用户设置以及功能控制有关的键信号输入。
马达191可以产生振动提示。马达191可以用于来电振动提示，也可以用于触摸振动反馈。例如，作用于不同应用(例如拍照，音频播放等)的触摸操作，可以对应不同的振动反馈效果。作用于显示屏194不同区域的触摸操作，马达191也可对应不同的振动反馈效果。不同的应用场景(例如：时间提醒，接收信息，闹钟，游戏等)也可以对应不同的振动反馈效果。触摸振动反馈效果还可以支持自定义。
指示器192可以是指示灯，可以用于指示充电状态，电量变化，也可以用于指示消息， 未接来电，通知等。
SIM卡接口195用于连接SIM卡。SIM卡可以通过插入SIM卡接口195，或从SIM卡接口195拔出，实现和电子设备100的接触和分离。电子设备100可以支持1个或N个SIM卡接口，N为大于1的正整数。SIM卡接口195可以支持Nano SIM卡，Micro SIM卡，SIM卡等。同一个SIM卡接口195可以同时插入多张卡。所述多张卡的类型可以相同，也可以不同。SIM卡接口195也可以兼容不同类型的SIM卡。SIM卡接口195也可以兼容外部存储卡。电子设备100通过SIM卡和网络交互，实现通话以及数据通信等功能。在一些实施例中，电子设备100采用eSIM，即：嵌入式SIM卡。eSIM卡可以嵌在电子设备100中，不能和电子设备100分离。
电子设备100的软件系统可以采用分层架构、事件驱动架构、微核架构、微服务架构或云架构。本申请实施例以分层架构对电子设备100的软件结构进行说明。
图3为本申请实施例提供的一种电子设备100的软件结构框图。
分层架构将软件系统分成若干个层，每一层都有清晰的角色和分工。层与层之间通过软件接口通信。在一些实施例中，分层架构将系统分为四层，从上至下分别为应用程序层，应用程序框架层，安卓运行时(Android runtime)和系统库，以及内核层。
应用程序层可以包括一系列应用程序包。
如图3所示，应用程序包可以包括相机，图库，日历，电话，地图，备忘录，通讯录，天气，音乐，视频，短信等应用程序。
应用程序框架层为应用程序层的应用程序提供应用编程接口(application programming interface，API)和编程框架。应用程序框架层包括一些预先定义的函数。
如图3所示，应用程序框架层可以包括个人知识图谱管理器，模型管理器等。
其中，个人知识图谱管理器用于构建用户的个人知识图谱。个人知识图谱管理器包括知识获取模块、知识融合模块、计算存储模块。
其中，知识获取模块用于获取个人数据。知识获取模块可以从两个方面来获取用户的个人数据。
一方面，电子设备中的各个应用程序在运行过程中产生的数据存放至数据服务和/或文件系统中，知识获取模块可以从数据服务和/或文件系统中获取到用户的个人数据。
另一方面，电子设备中的应用程序可以获得用户的授权，在获得用户授权之后，电子设备可以从各个应用程序获取到用户的个人数据。
之后，知识获取模块将个人数据发送至知识融合模块。
知识融合模块用于接收知识获取模块发送的个人数据，对个人数据进行预处理，并对预处理后的个人数据根据机器学习算法按照关系、事件、实体进行分类，得到用户的关系知识、事件知识、实体知识这三类知识。
用户知识包括关系知识、事件知识、实体知识。
关系知识是指根据个人数据获取到的用户的好友关系、同事关系、家人关系等人际关系的用户知识。
事件知识是指根据个人数据获取到的去旅行、去出差、去健身等已经发生的事情或正在进行的事情或未发生的事情的用户知识。
实体知识是指根据个人数据获取到的用户喜欢的电影、喜欢的音乐等实体的用户知识。
由上述实施例可知，知识获取模块获取的个人数据可以分为结构化的个人数据和非结构 化的个人数据。
对于非结构化的个人数据，知识融合模块可以将非结构化的个人数据(例如音频、视频、图片等)进行预处理。预处理包括将非结构化的个人数据(例如音频、视频、图片等)中包含的信息转换为文本信息。再对文本信息进行清洗。对文本信息进行清洗包括通过自然语言处理对文本信息进行分词和词性标注，根据预设词性删除掉文本信息中的不属于预设词性的词语，例如删掉冠词、介词、副词、连词、动词、语气词等。预设词性可以是用户自定义的，预设词性可以根据实际需要来设定。
之后，知识融合模块再对清洗后的文本信息进行去重。去重的目的在于，一方面，电子设备100从数据服务和/或文件系统中获取到用户的个人数据和从各个应用程序直接获取到用户的个人数据会出现重复的数据；另一方面，电子设备100直接从各个应用程序直接获取到用户的个人数据，各个应用程序中的个人数据也可能出现重复。因此，知识融合模块再对清洗后的文本信息进行去重，删除重复的数据，降低数据的冗余度。
其中，词性标注可以参考表1：
表1
Figure PCTCN2021117991-appb-000001
可以根据表1对文本信息进行分词后的词语进行词性标注。具体的，在形容词后面加上“/a”，在区别词后面加上“/b”，在连词后面加上“/c”，在副词后面加上“/d”，在叹词后面加上“/e”，在方位词后面加上“/f”，在语素后面加上“/g”，在前接成分后面加上“/h”，在成语后面加上“/i”，在简称后面加上“/j”，在后接成分后面加上“/k”，在习惯用语后面加上“/l”，在数词后面加上“/m”，在名词后面加上“/n”，在拟声词后面加上“/o”，在介词后面加上“/p”，在量词后面加上“/q”，在代词后面加上“/r”，在处所词后面加上“/s”，在时间词后面加上“/t”，在助词后面加上“/u”，在动词后面加上“/v”，在标点符号后面加上“/w”,在非语素字后面加上“/x”，在语气词后面加上“/y”，在状态词后面加上“/z”，在人名后面加上“/nr”，在地名后面加上“/ns”，在机构名称后面加上“/nt”，在其他专有名词后面加上“nz”。
示例性的，对于一个文本信息“《电视剧1》由演员1主演，将在2020年5月4日在苹果视频首播”。将该文本数据进行分词和词性标注后的结果是“《/w电视剧1/n》/w由/v演员1/nr主演/v，于/p2020/m年/m5/m月/m4/m日/m在/p苹果/n视频/n首播/v。/w”。
表1只是示例性的列举了一些词性标注的规则，还可以包括更多的词性标注规则，词性标注也可以参考其他的规则，本申请在此不做限定。
对于结构化的个人数据，知识融合模块可以将结构化的个人数据进行预处理，预处理包括数据清洗和去重两个步骤。对结构化的个人数据进行清洗和去重的方式与对非结构化的个人数据进行清洗和去重的方式是一样的，在此不在赘述。
将清洗和去重的个人数据输入到机器学习算法里面，利用机器学习算法将预处理后的个人数据按照关系知识、事件知识、实体知识进行分类。
其中，机器学习算法可以是基于关联规则的聚类算法。可以根据实际需求预先设定关联规则，然后根据设定的关联规则对预处理后的个人数据进行聚类分析处理，提取出预处理后的个人数据中的核心信息(关系知识、事件知识和实体知识)。该方法可以更加准确的提取出预处理后的个人数据中的与构建的知识图谱相关的核心信息(关系知识、事件知识和实体知识)。
同时，将预处理后的个人数据(第二个人数据)中与构建的知识图谱无关的核心信息统称为用户基本特征数据。用户基本特征数据可以包括：用户性别、年龄；用户使用的设备信息(例如设备标识、设备型号)等。
机器学习算法除了是基于关联规则的聚类算法，还可以是决策树分类法、朴素的贝叶斯分类算法、基于支持向量机的分类法等。本申请在此不做限定。
知识融合模块还用于将关系知识、实体知识、事件知识发送至计算存储模块。
计算存储模块用于接收知识融合模块发送的用户知识，并将用户知识按照预定结构存储起来。例如可以将用户知识按照五元组的结构存储起来。
具体的，计算存储模块可以将关系知识按照预定结构存储起来。将关系知识按照预定结构存储起来是为了将关系知识和时间对应存储起来，反映了关系知识与时间的联系。例如，可以按照五元组的结构式(第一五元组结构)将关系知识与时间存储起来，五元组的结构是“实体1-关系-实体2-时间点-时间区间”。
示例性的，我-客户-王总-2019.6.8-14。该关系知识表示我与客户王总于2019年6月8日认识，到今天(例如20202年8月10日)已经认识了16个月了。
计算存储模块可以将事件知识按照预定结构存储起来。将事件知识照预定结构存储起来是为了将事件知识和时间对应存储起来，反映了事件知识与时间的联系。例如，可以按照五元组的结构(第二五元组结构)将事件知识与时间存储起来，五元组的结构是事件-论元-逻辑关系-时间点-时间区间。这里，论元为支撑这个事件的动作，逻辑关系可以为因果关系、顺承关系等。
示例性，去出差-买飞机票-顺承-2020.9.1-3天。该事件知识表达了出差事件，出差与买飞机票是顺承关系，出发日期是2020年9月1日，行程安排是三天。
计算存储模块可以将实体知识按照预定结构存储起来。将实体知识按照预定结构存储起来是为了将实体知识和时间对应存储起来，反映了实体知识与时间的联系。例如，可以按照五元组的结构(第三五元组结构)将实体知识与时间存储起来，五元组的结构为“实体1：时间-关联权重-实体2-关系权重-实体3”。
示例性的，《电视剧1》：2020.4.6-1.0-演员1-0.8-演员3。该实体知识表达了《电视剧1》上映的时间是2020年4月6日，《电视剧1》与演员1的关联程度为1.0，演员1与演员3的关联程度为0.8。说明《电视剧1》与演员1的关联程度更高。
计算存储模块还用于根据预定结构的用户知识构建个人知识图谱。个人知识图谱是以图 形的形式来展示预定结构的用户知识的。可以参见图8所示的以图的形式表示构建的个人知识图谱。
计算存储模块还用于将用户的个人知识图谱发送至个人知识图谱库进行存储。
模型管理器包括特征池和模型池。其中，特征池用于存储用户的个人知识图谱和用户基础特征数据；模型池预置有重排序模型，重排序模型可以是逻辑回归、决策树、因子分解机(Factorization Machine，FM)、场感知分解机(Field-aware Factorization Machine，FFM)、深度学习等算法。模型池根据用户的个人知识图谱和用户基础特征数据等训练重排序模型。训练好的重排序模型(第一模型)可以对粗排广告进行进一步的筛选，得到投放广告。
下面对模型池如何训练重排序模型进行说明。
重排序模型的训练数据的输入是历史广告信息、用户基本特征数据、个人知识图谱，训练数据的输出是用户历史行为。
重排序模型可以是但不仅限于逻辑回归、决策树、FFM、深度学习等算法等。
模型池中存储的有以往显示的多个广告的历史广告信息，历史广告信息可以是广告的ID、广告的描述和大小等信息。
用户基本特征数据可以包括：用户性别、年龄、用户使用的设备信息(例如设备标识、设备型号)等。
用户历史行为可以是用户在过去一周时间内用户观看了哪些广告、用户观看广告的时长、用户关闭了哪些广告等。
将训练数据的输入丢入重排序模型，重排序模型将输出一个结果，该结果可以是用户点击并浏览了广告或用户未浏览并关闭了广告。将该结果与训练数据的输出作比较，若该结果与训练数据的输出不符，修改重排序模型的参数，继续训练重排序模型。当重排序模型输出的结果符合用户历史行为，则模型训练结束。重排序模型训练完成之后，重排序模型可以对粗排广告中的每个广告进行点击概率预测，即得到每个广告用户点击的概率值，重排序模型将广告按照用户点击的概率值从大到小进行排序，得到投放广告。
例如，可以根据用户的个人知识图谱中记载的关系知识可以获取到用户的人际关系网络，并根据人际关系网络学习到用户是一个善于社交的人还是内向的人。根据用户的个人知识图谱中记载的事件知识可以获取到用户的经常做的事情，例如旅游、出差等事件。根据用户的个人知识图谱中记载的实体知识可以获取到用户喜欢看的电视剧等。
示例性的，重排序模型根据用户的个人知识图谱学习到该用户是一个喜欢观看影视剧、偶尔会出去出差的用户。粗排广告包括一个购票类广告、一个影视剧类广告、一个交友类广告。重排序模型对粗排广告进行概率预测，得到购票类广告用户点击的概率是0.5，影视剧类广告用户点击的概率是0.9，交友类广告用户点击的概率是0.4。重排序模型根据用户点击的概率将粗排广告进行重新排序，得到投放广告。投放广告的排列顺序是影视剧类广告、购票类广告、交友类广告。
应用程序获取到重排序模型，应用程序可以根据训练好的重排序模型(第一模型)对粗排广告进行筛选，得到投放广告。投放广告是按照用户个人行为特征对粗排广告中的广告再次筛选序得到的，投放广告是按照用户个人可能点击并观看的广告的概率从高到低排序得到的。
下面对应用程序如何对粗排广告进行筛选得到投放广告进行说明。
在一些实施例中，应用程序得到粗排广告之后，应用程序获取到重排序模型，并根据重排序模型对粗排广告按照用户个人可能点击并观看的广告的概率从高到低排序，得到投放广 告。
在一些实施例中，应用程序得到粗排广告之后，应用程序获取到重排序模型，应用程序可以只保留用户个人可能点击并观看的广告的概率最高的一个广告。
在另一些实施例中，应用程序得到粗排广告之后，应用程序获取到重排序模型，应用程序可以将用户个人可能点击并观看的广告概率低于阈值的广告滤除掉。
在另一些实施例中，应用程序得到粗排广告之后，应用程序获取到重排序模型，并根据重排序模型对粗排广告按照用户个人可能点击并观看的广告的概率从高到低排序，应用程序还会判断各个广告是否在一定时间(例如三天)内已经推送过，若应用程序判断各个广告中的一些广告在一定时间(例如三天)内已经推送过，应用程序将滤除掉在一定时间(例如三天)内已经推送过的广告。
需要说明的是，对应用程序对粗排广告进行筛选得到投放广告还可以是其他的方式，本申请在此不做限定。
应用程序层中的模型管理器可以提供一个接口，应用程序服务器向广告服务器发送广告推荐请求，之后，应用程序接收广告服务器发送的粗排广告，应用程序可以获得用户的授权通过模型管理器提供的接口获取到重排序模型，应用程序通过重排序模型对粗排广告进行筛选，得到投放广告。应用程序将获取到的投放广告推荐给用户观看。
在一些实施例中，系统库中的个人知识图谱库可以提供一个接口，应用程序获得用户的授权后，应用程序可以从该接口获取到用户的个人知识图谱。应用程序可以通过用户的个人知识图谱为用户推荐用户感兴趣的内容。
示例性的，新下载的应用程序还未记录用户的行为，该新下载的应用程序可以获得用户的授权来获取到用户的个人知识图谱。该新下载的应用程序可以通过用户的个人知识图谱为用户推荐用户感兴趣的内容。一方面，不需要用户选择自己感兴趣的内容，新下载的应用程序可以根据个人知识图谱进行个性化的推荐。另一方面，新下载的应用程序可以根据个人知识图谱进行个性化的推荐，推荐的内容符合用户的行为特征，优化了内容推荐效果。
Android Runtime包括核心库和虚拟机。Android runtime负责系统的调度和管理。
核心库包含两部分：一部分是java语言需要调用的功能函数，另一部分是系统的核心库。
应用程序层和应用程序框架层运行在虚拟机中。虚拟机将应用程序层和应用程序框架层的java文件执行为二进制文件。虚拟机用于执行对象生命周期的管理，堆栈管理，线程管理，安全和异常的管理，以及垃圾回收等功能。
系统库可以包括多个功能模块。例如：数据服务、文件系统、个人知识图谱库，表面管理器(surface manager)，三维图形处理库(例如：OpenGL ES)等。
数据服务用于存储电子设备运存应用程序过程中产生的涉及用户隐私的结构化数据，例如数据库、表格等数据。
文件系统用于存储电子设备运存应用程序过程中产生的涉及用户隐私的非结构化数据、例如文档、图片、视频等数据。
个人知识图谱库用于存储个人知识图谱。
表面管理器用于对显示子系统进行管理，并且为多个应用程序提供了2D和3D图层的融合。
三维图形处理库用于实现三维图形绘图，图像渲染，合成，和图层处理等。
内核层是硬件和软件之间的层。内核层至少包含显示驱动，摄像头驱动，音频驱动，传 感器驱动。
参考图4，图4是本申请实施例提供的广告服务器200的硬件结构示意图。
广告服务器200可包括：一个或多个处理器301、存储器302、通信接口303、发射器305、接收器306、耦合器307和天线308。这些部件可通过总线304或者其他方式连接，图3以通过总线连接为例。其中：
通信接口303可用于广告服务器200与其他通信设备，其他通信设备可以是例如上述的电子设备或其他网络设备。示例性的，应用程序服务器发送广告推荐请求至广告服务器200，广告服务器200接收并响应应用程序服务器发送的广告推荐请求，广告服务器200发送粗排广告至电子设备100中的应用程序。具体的，通信接口303可以是长期演进(LTE)(4G)通信接口。不限于无线通信接口，广告服务器200还可以配置有有线的通信接口303来支持有线通信，例如广告服务器200与其他通信设备之间的回程链接可以是有线通信连接。
在本申请的一些实施例中，发射器305和接收器306可看作一个无线调制解调器。发射器305可用于对处理器301输出的信号进行发射处理。接收器306可用于接收信号。在广告服务器200中，发射器305和接收器306的数量均可以是一个或者多个。天线308可用于将传输线中的电磁能转换成自由空间中的电磁波，或者将自由空间中的电磁波转换成传输线中的电磁能。耦合器307可用于将移动通信号分成多路，分配给多个的接收器306。
存储器302与处理器301耦合，用于存储各种软件程序和/或多组指令。具体的，存储器302可包括高速随机存取的存储器，并且也可包括非易失性存储器，例如一个或多个磁盘存储设备、闪存设备或其他非易失性固态存储设备。
存储器302可以存储操作系统(下述简称系统)，例如uCOS、VxWorks、RTLinux等嵌入式操作系统。存储器302还可以存储网络通信程序，该网络通信程序可用于与一个或多个附加设备，一个或多个终端设备，一个或多个网络设备进行通信。
本申请实施例中，处理器301可用于读取和执行计算机可读指令。具体的，处理器301可用于调用存储于存储器302中的程序，例如本申请的一个或多个实施例提供的方法在广告服务器200侧的实现程序，并执行该程序包含的指令。
需要说明的，图4所示的广告服务器200的硬件结构仅仅是本申请实施例的一种实现方式，实际应用中，广告服务器200还可以包括更多或更少的部件，这里不作限制。
参考图5，图5是本申请实施例提供的另一种广告推荐系统架构示意图。
如图5所示，电子设备100可以包括知识获取模块5101、知识融合模块5102、计算存储模块5103、重排序模块5104。
对于知识获取模块5101、知识融合模块5102、计算存储模块5103、重排序模块5104各个模块的功能介绍，请参见图3所示的实施例，本身请在此不再赘述。
广告服务器200可以包括：广告点击模块5201、动态广告池模块5202、广告粗排模块5203、群体画像模块5204。
其中，广告点击模块5201，可用于接收应用程序服务器发送的广告推荐请求，响应于应用程序服务器发送的广告推荐请求，广告点击模块5201发送广告竞价请求至动态广告池模块5202。
动态广告池模块5202，可用于接收并响应广告点击模块5201发送的广告竞价请求。
一方面，动态广告池模块5202可以向多个广告商服务器发送获取广告请求，多个广告商服务器接收并响应获取广告请求，多个广告商服务器发送广告至动态广告池模块5202。另一 方面，动态广告池模块5202对多个广告商服务器发送的广告进行筛选(例如按照广告的价格从高到低进行筛选)，得到广告集，动态广告池模块5202发送广告集至广告粗排模块5203。
广告粗排模块5203，可用于接收动态广告池模块5202发送的广告集，广告粗排模块5203还用于接收群体画像模块5204发送的群体画像，广告粗排模块5203根据群体画像对广告集中的广告进行筛选，得到粗排广告，粗排广告可以包括一个或多个广告，广告粗排模块5203将粗排广告发送至应用程序。
电子设备100中的应用程序接收广告粗排模块5203发送的粗排广告，并获取重排序模型。应用程序根据重排序模型对粗排广告进行筛选，得到投放广告。应用程序将投放广告推荐给用户观看。
其中，群体画像模块5204可用于获取应用程序服务器发送的用户群体的群体数据，那么群体数据可以是用户群体点击了哪些广告、用户群体浏览了哪些广告、用户群体关闭了哪些广告、用户群体浏览广告的时长等。群体画像模块5204根据用户群体的群体数据为用户群体进行群体画像，群体画像即根据用户群体的群体数据为用户群体贴上标签，例如用户群体喜欢观看哪一类型的广告、用户群体不喜欢观看哪一类型的广告等。
下面结合电子设备上的用户界面，对电子设备直接从应用程序中获取用户的个人数据进行说明。
示例性的，如图6A所示，图6A示例性示出了电子设备上的通讯录应用的示例性用户界面60。
用户界面40可包括状态栏400和通讯录列表410。
可以按照朋友、同事、亲人等属性将用户联系人进行分类。如图6A所示，朋友类的联系人可以包括王可，王可的电话号码是12345；朋友类的联系人还包括李可，李可的电话是23456。同事类的联系人可以包括张三，张三的电话号码是12346；同事类的联系人还包括老王，老王的电话是12045；同事类的联系人可以包括张三，张三的电话号码是12365；同事类的联系人还包括小李，小李的电话是23045；同事类的联系人还包括花花，花花的电话是92345。亲人类的联系人可以包括妈妈，妈妈的电话是65430。
下面结合电子设备上的用户界面，举例介绍电子设备运行短信应用的过程当中产生的个人数据。
短信应用记载的有与用户相关的短信信息，电子设备可以从短信信息中获取到与用户相关的行为信息。例如用户通过购票应用程序购买了一张从北京到上海的机票，则购票应用程序发送用户的购票短信至短信应用，电子设备可以从短信应用获取到用户经常去的地方。又例如用户通过应用程序购买了一张电影票，则该应用程序发送用户的购票短信至短信应用，电子设备可以从短信应用获取到用户喜欢看的电影类型。
可以理解的是，电子设备也可以从购票应用程序直接获取到用户的购票短信。本申请在此不做限定。
示例性的，如图6B所示，图6B示例性示出了电子设备上的短信应用的示例性用户界面50。
用户界面50可包括状态栏400和短信内容显示框420。
短信内容显示框420可包括短信内容，短信内容是“订单EK123456,2020年8月24日Z234次10车24号下铺，深圳站12:45开，到达站为北京西。请持购票证件进站乘车。您购买的是电子客票，请直接持购票证件进展检票候车。”
电子设备可以根据短信内容可知该用户将于2020年8月24日乘火车从深圳站到北京西站，从而电子设备将酒店住宿类广告推荐给该用户观看。
下面结合电子设备上的用户界面，举例介绍电子设备运行通讯类应用的过程当中产生的个人数据。
图6C示例性示出了电子设备上的通讯类应用的示例性用户界面。
通讯类应用可以是电子设备预置的电话应用程序。通讯类应用也可以是用户从应用商店下载的通讯类应用程序，用户可以通过该通讯类应用程序与好友进行语音通话或视频通话。本申请在此不做限定。
下面以通讯类应用是电子设备预置的电话应用程序进行说明。
电子设备运行电话应用程序时，可以根据用户的通话记录获取用户联系人与用户的通话频率，从而电子设备可以根据用户联系人与用户通话频率推断出该用户联系人与用户的关系亲密度。
示例性的，如图6C所示，图6C示例性示出了电子设备上的电话应用程序的示例性用户界面40。
用户界面40可包括状态栏400和通话记录列表410。
状态栏400可包括时间指示符4001、电池状态指示符4002、无线高保真(wireless fidelity，Wi-Fi)信号的一个或多个信号强度指示符4003、移动通信信号(又可称为蜂窝信号)的一个或多个信号强度指示符4004。
通话记录列表410可包括一条或多条通话记录，例如今天上午十点二十二分，用户与张三打了一次电话，今天上午十一点三十分，用户与老王打了一次电话，今天下午一点十五分，用户与花花打了一次电话，今天下午一点二十一分，用户与妈妈打了一次电话，今天下午两点十一分，用户与王总打了一次电话，今天下午六点十分，用户与王可打了一次电话，今天下午六点十分，用户与李可打了一次电话，今天晚上八点十五分，用户与妈妈打了一次电话，昨天用户与妈妈打了一次电话。通话记录列表410还可以包括更多或更少的通话记录，在此不在举例。
下面结合电子设备上的用户界面，举例介绍电子设备运行备忘录应用的过程当中产生的个人数据。
备忘录应用中记载的有重要的事情以及时间。电子设备可以从备忘录应用中记载的重要的事情以及时间获取到与用户相关的重要的事情和人。例如好友生日，会议的时间、地点和内容等。示例性的，如图6D所示，图6D示例性示出了电子设备上的备忘录应用的示例性用户界面70。
用户界面40可包括状态栏400和我的备忘录列表440。
我的备忘录列表440可包括用户记载的重要的事情、重要的人和时间。例如，“8月7号下午两点工作汇报，准备PPT”，“王可8月16号生日，记得买生日礼物”，“8月20号下午三点的航班，到北京出差”。我的备忘录列表440还可以记载更多或更少的重要的事情、重要的人和时间，在此不再赘述。
下面结合电子设备上的用户界面，举例介绍电子设备运行备忘录应用的过程当中产生的个人数据。
天气应用中记载的有用户预设值城市的天气情况。电子设备可以从天气模块2106中获取到用户在的城市、该城市的天气情况等。
示例性的，如图6E所示，图6E示例性示出了电子设备上的天气的示例性用户界面80。
用户界面80可包括状态栏400和一周内天气情况列表450。
一周内天气情况列表450可包括一周内某地点的每天天气情况。例如，电子设备的所在地 为深圳市南山区，今天(星期二)当时的天气为阴天，温度为29°，今天(星期二)晚上九点的天气为阴天，温度为28°，今天(星期二)晚上十点的天气为阴天，温度为28°，今天(星期二)晚上十一点的天气为阴天，温度为27°，明天(星期三)凌晨零点的天气为阴天，温度为27°，明天(星期三)凌晨一点的天气为阴天，温度为27°。星期三的天气为阵雨，最高温度为29°，最低温度为27°；星期四的天气为阵雨，最高温度为29°，最低温度为27°；星期五的天气为阵雨，最高温度为29°，最低温度为27°；星期六的天气为阵雨，最高温度为29°，最低温度为27°；星期天的天气为阵雨，最高温度为29°，最低温度为27°；下个星期一的天气为阵雨，最高温度为29°，最低温度为27°。
上述UI实施例只是示例性说明电子设备100可以从上述应用程序获取到用户的个人数据。个人数据还可以来自于其他的应用程序，本申请在此不在一一介绍。
如图7所示，图7为本申请实施例提供的一种广告显示方法流程示意图。
该方法包括：
S701、电子设备100获取个人数据，并对个人数据进行预处理。
一方面，电子设备中的各个应用程序在运行过程中产生的数据存放至数据服务和/或文件系统中，电子设备100可以从数据服务和/或文件系统中获取到用户的个人数据。
另一方面，电子设备中的应用程序可以获得用户的授权，在获得用户授权之后，电子设备可以直接从各个应用程序获取到用户的个人数据。
示例性的，电子设备100可以从通讯录应用、通讯类应用等直接获取到用户的人际关系知识，例如，用户的朋友关系、用户的同事关系、用户的亲人、用户的客户关系等用户知识。
电子设备100可以从短信应用、备忘录运用等直接获取到用户的好友生日、会议信息、旅游城市、学术论文、出发日期等用户知识。
电子设备100可以从图库应用中获取到用户喜欢的事物、喜欢的城市、喜欢的歌手、歌名、感兴趣的演员等用户知识。
电子设备100可以从可穿戴设备(例如蓝牙手表)中获取到用户所在的城市区域、天气情况、运动状态等用户知识。
示例性的，电子设备100从视频应用中获取到对《电视剧1》的描述信息和数据。《电视剧1》的描述信息和数据如表2所示：
表2
Figure PCTCN2021117991-appb-000002
电子设备100还用于从个人数据进行预处理。预处理包括对个人数据进行清洗和去重。 对个人数据进行清洗包括通过自然语言处理对文本信息进行分词和词性标注，根据预设词性删除掉文本信息中的不属于预设词性的词语，例如删掉冠词、介词、副词、连词、动词、语气词等。预设词性可以是用户自定义的，预设词性可以根据实际需要来设定。
之后，一方面，电子设备100从数据服务和/或文件系统中获取到用户的个人数据和从各个应用程序直接获取到用户的个人数据会出现重复的数据；另一方面，电子设备100直接从各个应用程序直接获取到用户的个人数据，各个应用程序中的个人数据也可能出现重复。因此，知识融合模块再对清洗后的文本信息进行去重，删除重复的数据，降低数据的冗余度。
示例性的，电子设备100对获取到的《电视剧1》的描述信息和数据进行预处理，具体包括对《电视剧1》的描述信息和数据进行分词、词性标注。词性标注的规则参考图3实施例，本申请在此不在赘述。
电子设备100对《电视剧1》的描述信息和数据进行预处理的结果如表3所示：
表3
Figure PCTCN2021117991-appb-000003
示例性的，电子设备100对分词和词性标注后的《电视剧1》的描述信息和数据按照预设的词性进行清洗,得到包括人名、地名、歌名、影视剧名、时间、演员等词语。对分词和词性标注后的《电视剧1》的描述信息和数据如表4所示：
表4
Figure PCTCN2021117991-appb-000004
S702、电子设备100对预处理后的个人数据根据机器学习算法按照关系、事件、实体进行分类，得到用户的关系知识、事件知识、实体知识这三类知识。
具体的，参见图3实施例，在此不再赘述。
S703、电子设备100将用户知识按照预定结构存储起来，并根据预定结构的用户知识构建个人知识图谱。
具体的，电子设备100将关系知识按照预定结构存储起来。将关系知识按照预定结构存储起来是为了将关系知识和时间对应存储起来，反映了关系知识与时间的联系。例如，可以按照五元组的结构式将关系知识与时间存储起来，五元组的结构是“实体1-关系-实体2-时间点-时间区间”。
示例性的，我-客户-王总-2019.6.8-14。该关系知识表示我与客户王总于2019年6月8日认识，到今天(例如20202年8月10日)已经认识了16个月了。
这里，时间区间用月数表示203天的，也可以用年数表示，也可以用天数表示，本申请自此不做限定。
在一些实施例中，电子设备100也可以仅将实体之间的关系仅用时间点表示。基本格式是：实体1-关系-实体2-时间点。
示例性的，我-同事-张三-2019.4.5。该关系知识表示我与同事张三于2019年4月15日认识。
电子设备100将事件知识按照预定结构存储起来。将事件知识照预定结构存储起来是为了将事件知识和时间对应存储起来，反映了事件知识与时间的联系。例如，可以按照五元组的结构将事件知识与时间存储起来，五元组的结构是“事件-论元-逻辑关系-时间点-时间区间”。
这里，论元为支撑这个事件的动作，逻辑关系可以为因果关系、顺承关系等。
示例性，去出差-买飞机票-顺承-2020.9.1-3天。该事件知识表达了出差事件，出差与买飞机票是顺承关系，出发日期是2020年9月1日，行程安排是三天。
示例性，去免税店-买化妆品-因果-2020.8.20-2天。该事件知识表达了购买事件，去免税店与买化妆品是因果关系，出发日期是2020年8月20日，行程安排是两天。
在一些实施例中，电子设备100也可以仅将用户发生的事件用时间点表示。基本格式是：实体1：事件-论元-逻辑关系-时间点。
示例性，去出差-买飞机票-顺承-2020.9.1。该事件知识表达了出差事件，出差与买飞机票是顺承关系，出发日期是2020年9月1日。
电子设备100将实体知识按照预定结构存储起来。将实体知识按照预定结构存储起来是为了将实体知识和时间对应存储起来，反映了实体知识与时间的联系。例如，可以按照五元组的结构将实体知识与时间存储起来，五元组的结构为“实体1：时间-关联权重-实体2-关系权重-实体3”。
示例性的，《电视剧1》：2020.4.6-1.0-演员1-0.8-演员3。该实体知识表达了《电视剧1》上映的时间是2020年4月6日，《电视剧1》与演员1的关联程度为1.0，演员1与演员3的关联程度为0.8。说明《电视剧1》与演员1的关联程度更高。
示例性的，《电视剧2》：2020.2.1-0.6-演员2-0.8-演员1。该实体知识表达了《电视剧2》上映的时间是2020年2月1日，《电视剧2》与演员2的关联程度为0.6，演员2与演员1的关联程度为0.8。说明《电视剧2》与演员1的关联程度更高。
示例性的，电子设备100对《电视剧1》清洗和去重后的描述信息和数据进行实体挖掘， 并建立基于时间的实体知识，电视剧1：2020-1.0-演员1-0.8-演员2。
电子设备100根据预定结构的事件知识、预定结构的关系知识、预定结构的实体知识构建个人知识图谱。
个人知识图谱就是以图的形式表示预定结构的用户知识的。
如图8所示，图8为本申请实施例中以图的形式表示构建的个人知识图谱的示意图。可以理解的是，图8中的个人知识图谱只显示了用户部分用户知识的个人知识图谱，该个人知识图谱还可以包括更多或更少的用户知识。
如图8所示，该用户的个人知识图谱可以包括关系知识、事件知识和实体知识。
对于图8中所示的个人知识图谱对用户的关系知识、事件知识和实体知识的具体描述，参见上述实施例所述，本申请再次不再赘述。
需要说明的是，由于电子设备100从各个应用程序模块中获取的用户知识是在实时变化的，因此电子设备100根据从各个应用程序模块中获取的用户知识构建个人知识图谱也是需要实时更新的，这样，个人知识图谱会更准确的表达用户的特征。
电子设备100构建了用户基于时间的个人知识图谱之后，电子设备100还可以更新个人知识图谱中的用户知识。具体个人知识图谱可以只保留用户最近一段时间的用户知识，电子设备100可以把个人知识图谱中不属于最近一段时间的用户知识滤除掉，一方面，更新后的个人知识图谱刻画的用户特征就更准确。另一方面，可以节省电子设备100的存储资源。
下面对电子设备100如何更新个人知识图谱进行说明。
电子设备100更新个人知识图谱可以分为两个方面，第一方面：电子设备100将用户的新的知识加入到个人知识图谱中；第二方面：电子设备100将个人知识图谱中用户已存在的知识滤除掉。
下面对电子设备100滤除掉个人知识图谱中用户已存在的知识进行说明。
具体的，在一些实施例中，电子设备100可以只保留个人知识图谱中最近一段时间(例如两年)的用户知识，将个人知识图谱中最近一段时间(例如两年)以外的用户知识滤除掉。
在另一些实施例中，电子设备100还可以根据个人知识图谱的内存大小来滤除掉个人知识图谱中用户已存在的知识。
具体的，电子设备100检测到用户的个人知识图谱的内存大小接近预设的内存大小，那么电子设备100可以将根据预设的内存大小保留个人知识图谱中最近一段时间的知识。
本申请实施例还可以是其他的方式来更新用户的个人知识图谱，上述实施例仅用于解释本申请，不应构成限定。
电子设备100可以在特定时间区间更新个人知识图谱。
在一些实施例中，电子设备100可以按照固定时间(例如一天)更新个人知识图谱。
在另一些实施例中，电子设备100可以根据用户的行为习惯更新个人知识图谱。示例性的，在时间段“22:00-8:00”之间，用户在家休息，电子设备100处于待机状态，电子设备100可以在时间段“22:00-8:00”之间更新个人知识图谱。因为电子设备100更新个人知识图谱时会占用电子设备100一定的内存空间。这样，电子设备100不在用户休息时间更新个人知识图谱，为用户操作余留更多的内存空间，保证了用户操作的流畅性。
S704、电子设备100根据历史广告信息、用户基本特征数据、个人知识图谱、用户历史行为训练重排序模型。
对于电子设备100如何训练重排序模型，参见图3实施例，在此不再赘述。
需要说明的是，电子设备100是根据个人知识图谱等数据来训练重排序模型的，由上述 实施例可以知道个人知识图谱是会随着时间来更新的，那么电子设备100也应该随着时间来更新重排序模型，这样重排序模型对输入的多个广告进行重新排序得到的广告排序列表就更准确。
下面对电子设备100如何更新重排序模型行说明。
在一些实施例中，电子设备100可以按照固定时间(例如一天)更新重排序模型。
在一些实施例中，电子设备100可以根据用户的行为习惯更新个人知识图谱。
示例性的，在时间段“22:00-8:00”之间，用户在家休息，电子设备100处于待机状态，电子设备100可以在时间段“22:00-8:00”之间更新重排序模型。因为电子设备100更新重排序模型时会占用电子设备100一定的内存空间，这样，电子设备100不在用户使用电子设备100期间更新重排序模型，为用户操作余留更多的内存空间，保证了用户操作的流畅性。
S705、电子设备100中的应用程序接收广告服务器200发送的粗排广告，
电子设备100检测到用户操作，响应于用户操作，电子设备100发送用户请求至应用程序服务器300，应用程序服务器300发送广告推荐请求至广告服务器200。
该用户操作例如可以是：电子设备100检测到用户开启了应用程序，或者电子设备100检测到用户下一个视频了应用程序的当前浏览页面(例如电子设备100检测到用户单指向下滑动来下一个视频当前浏览界面)，则电子设备100发送用户请求至应用程序服务器300。
示例性的，如图9A所示，图9A示例性示出了应用程序的当前浏览界面的示例性用户界面90。
用户界面90可包括状态栏400、当前视频浏览界面460、广告显示区域470。
当前视频浏览界面460包括推荐控件4601、热门控件4602、小视频控件4603、下一个视频控件4604、视频显示窗口4605、点赞控件4606、评论控件4607、转发控件4608。
广告显示区域470包括广告图标4609、关闭控件4610、广告链接控件4611。
其中，视频显示窗口4605显示的是当前播视视频(例如电视剧1)的视频内容4600。
下一个视频控件4604可以接收用户单击操作，响应于用户的单击操作，当前视频浏览界面460将显示其他视频的视频内容。
响应于用户的单击下一个视频控件4604操作，电子设备100发送用户请求至应用程序服务器300，应用程序服务器300发送广告推荐请求至广告服务器200，响应于广告推荐请求，广告服务器200将发送新的广告至电子设备100中的应用程序，广告显示区域470将显示新的广告内容。
应用程序的当前浏览界面也可以接受用户单指向下滑动来下一个视频当前浏览界面，响应于用户的作用于当前浏览界面的单指向下滑动操作，电子设备100发送用户请求至应用程序服务器300，应用程序服务器300发送广告推荐请求至广告服务器200，响应于广告推荐请求，广告服务器200将发送新的广告至电子设备100中的应用程序，广告显示区域470将显示新的广告内容。
下面对广告服务器200如何将粗排广告发送至电子设备100中的应用程序进行说明。
首先，广告服务器200根据群体数据为用户群体进行群体画像。
广告服务器200获取群体数据，群体数据可以是用户点击了哪些广告、用户浏览了哪些广告、用户关闭了哪些广告等。
图9B-图9C示出了广告服务器200获取群体数据的示例性用户界面的UI图。
示例性的，如图9B所示，图9B示例性示出了应用程序的当前浏览界面的示例性用户界 面900。
用户界面900可包括状态栏400、当前视频浏览界面460、广告显示区域470。
当前视频浏览界面460包括推荐控件4601、热门控件4602、小视频控件4603、下一个视频控件4604、视频显示窗口4605、点赞控件4606、评论控件4607、转发控件4608。
广告显示区域470包括广告图标4609、关闭控件4610、广告链接控件4611。
其中，视频显示窗口4605显示的是当前播视视频(例如电视剧1)的视频内容4600。
下一个视频控件4604可以接收用户单击操作，响应于用户的单击操作，当前视频浏览界面460将显示其他视频的视频内容。
广告显示区域470显示的有广告内容，广告内容是“领劵方式已经放在下方了，请点击领取ABCDEFGHI”。
其中，广告链接控件4611可以接收用户单击操作，响应于用户的单击操作，用户界面910将显示该广告的用户界面。同时，响应于用户单击广告链接控件4611的操作，应用程序服务器300将用户点击并观看了这个广告的行为上报至广告服务器200。用户点击并观看了这个广告的行为可以作为该广告的一份业务数据。
关闭控件4610也可以接收用户单击操作，响应于用户的单击操作，广告显示区域470显示如图9C所示的选择提示框4612，选择提示框4612包括不感兴趣控件4613、重复推荐控件4614、屏蔽该类广告控件4615。
其中，不感兴趣控件4613可以接收用户单击操作，响应于用户的单击操作，广告显示区域470将不在显示该广告；重复推荐控件4614可以接收用户单击操作，响应于用户的单击操作，广告显示区域470将在一定时间(例如48小时)内不在推荐该广告给用户观看；屏蔽该类广告控件46154614可以接收用户单击操作，响应于用户的单击操作，广告显示区域470不会推荐该类(例如美妆类)广告给用户观看。
同时，响应于用户单击关闭控件4610的操作，应用程序服务器300将用户关闭并未观看这个广告的行为上报至广告服务器200。用户关闭并未观看这个广告的行为可以作为该广告的一份群体数据。
该广告的群体数据还可以来源于其他的途径，上述实施例仅用于解释本申请，本申请实施例在此不做限定。
应用程序服务器300收集群体数据并上报至广告服务器200。广告服务器200根据所有用户的群体为用户群体进行群体画像，群体画像即根据所有用户的群体数据为用户群体贴上标签，例如用户群体喜欢观看哪一类型的广告、用户群体不喜欢观看哪一类型的广告等。
可以理解的是，应用程序服务器300收集所有用户的群体数据是实时变化的，因此广告服务器200接收应用程序服务器300发送的所有用户的群体数据也是在实时变化的。
广告服务器200可以根据所有用户的群体数据按照固定时间(例如一天)为用户群体进行群体画像。那么用户群体的群体画像也是周期性的更新的。
之后，广告服务器200根据用户群体的群体画像对广告进行筛选，得到粗排广告。
粗排广告可以为一个或多个广告。
粗排广告是根据用户群体的群体画像对广告集中的多个广告进行筛选得到的。广告服务器200将粗排广告发送至电子设备100中的应用程序。
电子设备100中的应用程序接收广告服务器200发送的粗排广告。
由上述实施例可知，在一些实施例中，广告服务器200也可以将用户群体在分类，例如可以将用户群体分为女性用户群体和男性用户群体，或者用户群体也可以分为各个年龄段的 用户群体等。
示例性的，当广告服务器200将用户群体分为女性用户群体和男性用户群体时，广告服务器200分别为男性用户群体进行画像和对女性用户群体进行画像。
具体的，广告服务器200根据男性用户的群体画像对多个广告进行筛选，得到男性用户的粗排广告。广告服务器200将男性用户的粗排广告推送至男性用户的电子设备上。
广告服务器200根据女性用户的群体画像对多个广告进行筛选，得到女性用户的粗排广告。广告服务器200将女性用户的粗排广告推送至女性用户的电子设备中的应用程序上。
示例性的，当广告服务器200将用户群体分为各个年龄段的用户群体时，群体画像还可以根据各个年龄段的用户群体对用户群体画像。
具体的，广告服务器200对用户年龄在21-35岁之间的用户群体进行画像，并根据年龄在21-35岁之间的用户的群体画像对多个广告进行筛选，得到年龄在21-35岁之间的用户的粗排广告。广告服务器200将年龄在21-35岁之间的用户的粗排广告推送至年龄在21-35岁之间的用户的电子设备中的应用程序上。
广告服务器200对用户年龄在36-50岁之间的用户群体进行画像，并根据年龄在36-50岁之间的用户的群体画像对多个广告进行筛选，得到年龄在36-50岁之间的用户的粗排广告。广告服务器200将年龄在36-50岁之间的用户的粗排广告推送至年龄在36-50岁之间的用户的电子设备中的应用程序上。
广告服务器200对用户年龄在51-70岁之间的用户群体进行画像，并根据年龄在51-70岁之间的用户的群体画像对多个广告进行筛选，得到年龄在51-70岁之间的用户的粗排广告。广告服务器200将年龄在51-70岁之间的用户的粗排广告推送至年龄在51-70岁之间的用户的电子设备中的应用程序上。
S706、电子设备100中的应用程序获取到重排序模型，应用程序根据重排序模型对粗排广告进行筛选，得到投放广告。
投放广告的数量可以为一个或多个广告。
对于应用程序如何根据粗排广告得到投放广告，参见图3实施例，在此不再赘述。
下面对应用程序如何显示投放广告进行说明。
电子设备100开启了应用程序，应用程序的当前浏览界面有一个广告位，该广告位可以显示一个或多个广告。
示例性的，该广告位的广告时长为60秒，那么该广告位可以仅显示一个广告，该广告的时间为60秒。
示例性的，该广告位的广告时长为60秒，那么该广告位可以显示6个广告，每个广告的显示时间为10秒。
需要说明的是，上述实施例仅用于解释本申请，本申请对于每个广告位显示的广告数量和广告时间不做限定。
本申请不限于广告推荐，还可以运用于内容推荐，内容推荐可以包括歌曲推荐、电子书推荐、影视剧推荐、美食推荐、购物推荐等等，本申请在此不做限定。
如图10所示，图10为本申请实施例提供的另一种广告显示方法流程示意图。
该方法包括：
S1001、电子设备100获取第一个人数据。
电子设备100获取第一个人数据，第一个人数据为用户的个人信息。个人信息可以是以下一项或多项：性别、年龄、性格、爱好、人际关系、收入、通讯录信息、通话记录、短信、备忘录信息、居住的地址、所述居住的地址的天气情况。
在一些实施例中，电子设备100可以每隔固定周期(例如一周)获取所述用户的第一个人数据。
具体的，电子设备100获取第一个人数据请参考图6A-图6D实施例以及S701所述的实施例，本申请再此不再赘述。
S1002、电子设备100根据第一个人数据构建个人知识图谱。
电子设备100在根据第一个人数据构建个人知识图谱之前，需要对第一个人数据进行预处理。电子设备100对第一个人数据进行预处理包括以下两个步骤：
步骤一、电子设备将第一个人数据转化为文本信息，并对文本信息进行断句、分词和词性标注。电子设备从文本信息中获取属于预设词性的词语。
步骤二、电子设备从文本信息中获取属于预设词性的词语之后，电子设备100对文本信息中的词语进行去重，去除数据冗余。
具体的，电子设备100需要获取文本信息中出现次数为一次的词语；当文本信息中有出现两个及两个以上相同的词语，电子设备100保留文本信息中有出现两个及两个相同词语中的一个词语。
即电子设备100从第一个人数据中获取到第二个人数据。第二个人数据包括关系知识、事件知识和实体知识。
具体的，电子设备100对第一个人数据进行预处理请参考S701所述的实施例，本身请在此不做限定。
电子设备100在根据第一个人数据构建个人知识图谱，具体包括以下步骤：
电子设备100将关系知识、事件知识和实体知识按照预定结构存储。预定结构可以是五元组结构。
具体的，电子设备100将关系知识按照第一五元组结构进行存储；第一五元组结构为“第一实体-关系-第二实体-第一时间点-第一时间区间”；关系表征第一实体与第二实体的关系，第一时间点为第一实体与第二实体建立关系的时间，第一时间区间为第一时间点到当前时间点的间隔时间。
电子设备100将事件知识按照第二五元组结构进行存储；第二五元组结构为“事件-论元-逻辑关系-第二时间点-第二时间区间”；论元为事件的发生动作，逻辑关系表征事件与论元的关系，第二时间点为事件发生的时间，第二时间区间为第二时间点到当前时间点的间隔时间。
电子设备100将实体知识按照第三五元组结构进行存储；第三五元组结构为“第三实体：第三时间点-第一关联权重-第四实体-第二关联权重-第五实体”；第三时间点为第三实体的发生时间，第一关联权重为第三实体与第四实体的关联程度，第二关联权重为第四实体与所述第五实体的关联程度。
电子设备根据预定结构的关系知识、预定结构的事件知识、预定结构的实体知识构建用户的个人知识图谱。
具体的，电子设备100构建个人知识图谱可以参考S703所述的实施例，本申请在此不再 赘述。
电子设备100还可以更新个人知识图谱。
一方面，电子设备100可以删除个人知识图谱中的个人数据，来更新个人知识图谱。
具体的，电子设备删除个人知识图谱中第一时间区间大于第一阈值的关系知识；和/或，电子设备删除个人知识图谱中第二时间区间大于第一阈值的事件知识；和/或，电子设备根据第三时间点确定出第三时间点到当前时间点的第三时间区间；电子设备删除个人知识图谱中第三时间区间大于第一阈值的实体知识。
另一方面，电子设备100可以添加新的个人数据到个人知识图谱中。
具体的，电子设备100每隔固定周期获取到第一个人数据，并将第一个人数据加入到个人知识图谱中。
S1003、电子设备100从广告服务器200获取第一广告内容的参数信息。
电子设备接收到广告服务器的可以是第一广告内容的参数信息。参数信息可以是第一广告内容的类型、链接地址、大小等信息。第一广告内容可以包括一个或多个广告。
第一广告内容是一下任意一种或几种：图片、视频、文字、音频等。第一广告内容可以包括一个或多个广告。
S1004、电子设备100根据所述个人知识图谱从第一广告内容的参数信息中获取到第二广告内容的参数信息。
电子设备根据个人知识图谱从第一广告内容的参数信息中获取到第二广告内容的参数信息可以采取以下方式中的一种或多种。方式一、电子设备保留第一广告内容的参数信息中的所有广告的参数信息，电子设备只是将广告的类型按照用户的喜爱度预测值从高到低将第一广告内容进行排序，得到第二广告内容的参数信息。方式二、电子设备从第一广告内容的参数信息中筛选一部分广告的参数信息得到第二广告内容的参数信息。具体的，电子设备将广告的类型按照用户的喜爱度预测值从高到低将第一广告内容进行排序，仅保留用户的喜爱度预测值高于第一阈值的广告的参数信息，得到第二广告内容的参数信息。
第二广告内容是一下任意一种或几种：图片、视频、文字、音频等。第二广告内容可以包括一个或多个广告。
具体的，电子设备100根据个人知识图谱训练重排序模型，得到第一模型，电子设备通过第一模型从第一广告内容的参数信息中获取到第二广告内容的参数信息。
具体的，电子设备100根据个人知识图谱训练重排序模型，可以包括以下内容：
电子设备获取到用户历史行为和电子设备显示的历史广告信息。电子设备将历史广告信息、个人知识图谱作为所述重排序模型的输入，重排序模型输出第一结果。电子设备将第一结果与用户历史行为比较，并修改重排序模型的参数，直至重排序模型输出的第一结果在预设范围内，得到第一模型。
S1005、电子设备100根据第二广告内容的参数信息获取到第二广告内容。
电子设备100根据第二广告内容的参数信息(例如链接地址)获取到第二广告内容。
S1006、电子设备100在显示屏上显示第二广告内容。
电子设备100在显示屏上显示所述第二广告内容，可以参考以下方式：
方式一：电子设备按照第二广告内容中用户的喜爱度预测值从高到低播放所述第二广告 内容中的一个或多个广告。
方式二：电子设备显示第二广告内容中用户的喜爱度预测值最高的广告。
方式三、电子设备按照第二广告内容中用户的喜爱度预测值从高到低播放第二广告内容中的一个或多个广告，并屏蔽掉第二广告内容中电子设备在第一时间段内播放过的一个或多个广告。
在电子设备100显示完第二广告内容之后，电子设备100可以获取用户对第二广告内容的观看数据；观看数据包括用户观看了第二广告内容中的一个或多个广告的广告类型和用户关闭了第二广告内容中的一个或多个广告的广告类型；
电子设备根据所述观看数据更新第一模型。这样，电子设备根据用户的观看广告的数据来更新第一模型，第一模型会在下一次给用户推荐广告时将用户观看次数最多的类型的广告推荐给用户，这样，更符合用户的需求。
在本申请以下实施例中，电子设备100根据用户的个人数据构建个人知识图谱之后，电子设备100中的应用程序可以征求用户的同意获取个人知识图谱，应用程序可以根据个人知识图谱为用户进行个性化的推荐。这样，一方面，应用程序只有征求用户的同意才能获取个人知识图谱，充分尊重了用户个人隐私；另一方面，应用程序根据个人知识图谱为用户进行个性化的推荐，这样，应用程序为用户推荐的内容更符合用户的行为特征。
首先，对目前已有的一种内容显示方法进行说明。
用户使用了电子设备100一段时间之后，用户从应用商店下载了一个新的应用程序(例如第一应用)，一方面，第一应用会提示用户注册个人信息并登陆，另一方面，第一应用可以提示用户选择自己感兴趣的内容，第一应用将用户感兴趣的相关的内容推荐至用户观看。
示例性的，下面结合附图对上述应用场景进行说明。
图11A示例性示出了电子设备100上的用于应用程序菜单的示例性用户界面700。
用户界面700可包括：状态栏400，具有常用应用程序图标的托盘710，导航栏720，以及其他应用程序图标。其中：
具有常用应用程序图标的托盘710可展示：电话图标7012、联系人图标7013、短信图标7014、相机图标7015。
导航栏720可包括：返回按键7016、主界面(Home screen)按键7017、呼出任务历史按键7018等系统导航键。当检测到用户点击返回按键7016时，电子设备100可显示当前页面的上一个页面。当检测到用户点击主界面按键7017时，电子设备100可显示主界面。当检测到用户点击呼出任务历史按键7018时，电子设备100可显示用户最近打开的任务。各导航键的命名还可以为其他，本申请对此不做限制。不限于虚拟按键，导航栏720中的各导航键也可以实现为物理按键。
其他应用程序图标可例如：时钟的图标7001、日历的图标7002、图库的图标7003、备忘录的图标7004、文件管理的图标7005、电子邮件的图标7006、音乐的图标7007、计算器的图标7008、华为视频的图标7009、运动健康的图标7010、第一应用的图标7011。
其中，第一应用是用户从应用商店下载后还未使用过的应用。第一应用可以是购物应用，可以是电子书应用程序、还可以是视频应用程序等等。该应用场景以第一应用程序是购物应用进行说明。
如图11A所示，第一应用图标7011可以接收用户的单击操作，响应于用户的单击操作，电子设备100显示如图11B所示的示例性用户界面730。
用户界面730包括状态栏400和性别选择界面740。
其中，性别选择界面740包括控件7101、控件7102。
控件7101可以接收用户单击操作，响应于用户的单击操作，第一应用将推荐男性感兴趣的物品至用户观看。
控件7102可以接收用户单击操作，响应于用户的单击操作，第一应用将推荐女性感兴趣的物品至用户观看。
示例性的，如图11B所示，控件7102接收用户单击操作，响应于用户的单击操作，电子设备100显示如图11C所示的示例性用户界面750。
用户界面750包括状态栏400和个性化推荐选择界面760。
其中，个性化推荐选择界面760包括多个推荐主题控件和下一步控件770。
多个推荐主题控件可以包括穿搭控件7501、运动控件7502、美妆控件7503、旅游控件7504、美食控件7505、游戏控件7506。多个推荐主题控件还可以包括其他的推荐控件，本身请在此不在限定。
多个推荐主题控件中的任意一个推荐主题控件可以接收用户单击选中，响应于用户的单击操作，第一应用程序将推荐用户感兴趣的主题至用户观看。
示例性的，用户对美妆、穿搭和美食感兴趣，那么穿搭控件7501、美妆控件7503、美食控件7505接收用户的单击选中，下一步控件770接收并响应于用户的单击操作，电子设备100将显示第一应用的示例性用户界面，第一应用的用户界面显示的内容为用户选择的美妆、穿搭和美食类内容。这样，第一应用可以根据用户的喜好为用户推荐相关的内容。
但是，从上述实施例可以知道，当用户打开了一个未使用过的应用程序，用户需要先选择性别，再选择用户感兴趣的主题内容，用户选择完之后，用户需在点击下一步才能进入到应用程序中。从上述操作可知，该应用程序的推荐方式操作复杂，用户体验不好。
第一应用可以基于本申请实施例提供的个人知识图谱来推荐用户感兴趣的内容，一方面，用户不需要执行一系列操作来选择自己感兴趣的内容，另一方面，第一应用根据个人知识图谱来推荐用户感兴趣的内容会更准确。
当用户打开了第一应用，第一应用将征求用户的请求来获取个人知识图谱等信息，这样，当用户授权给第一应用时，第一应用才会获取个人知识图谱等数据，这样，充分尊重了用户个人隐私。
如图12所示，图12为本申请实施例的另一种系统架构示意图。该系统包括电子设备100和应用服务器300。
其中，应用服务器300发送第一内容列表至电子设备100中的第一应用，第一内容列表中可以包括多个内容，第一应用接收应用服务器300发送的第一内容列表。
第一应用获取用户的请求后，电子设备100将个人知识图谱发送至第一应用，第一应用接收电子设备100发送的个人知识图谱，第一应用获取第一内容列表后，第一应用根据个人知识图谱对第一内容列表中的多个内容进行筛选，得到第二内容列表，第一应用将第二内容列表中的内容推荐给用户观看。
示例性的，如图12A所示，图12A为第一应用获取用户请求的示例性用户界面780。
用户界面780包括状态栏400和提示框790。
提示框790显示的有提示信息和不同意控件7801以及同意控件7802。其中，该提示信息用于提示用户是否同意第一应用获取个人知识图谱等信息。提示信息包括“为了更好的提 供浏览推荐、发布信息、购买商品等相关服务，我们会根据您使用服务的具体功能需要、收集必要的用户信息(可能涉及设备、个人知识图谱等信息)”。
不同意控件7801可以接受用户的单击操作，响应于用户的单击操作，用户不同意第一应用获取个人知识图谱等数据，那么第一应用将不会为用户推荐个性化内容。
同意控件7802可以接受用户的单击操作，响应于用户的单击操作，用户同意第一应用获取个人知识图谱等数据，那么第一应用将根据个人知识图谱等数据为用户推荐与用户的个人行为特征相近的内容，例如根据用户的喜好、消费水平等推荐类似的内容给用户观看。
这样，第一应用根据个人知识图谱等数据为用户推荐与用户的个人行为特征相近的内容，不需要用户手动去选择自己感兴趣的内容，一方面，用户操作简单，另一方面，第一应用根据个人知识图谱来推荐用户感兴趣的内容会更准确，更符合用户的需求。
上述实施例中所用，根据上下文，术语“当…时”可以被解释为意思是“如果…”或“在…后”或“响应于确定…”或“响应于检测到…”。类似地，根据上下文，短语“在确定…时”或“如果检测到(所陈述的条件或事件)”可以被解释为意思是“如果确定…”或“响应于确定…”或“在检测到(所陈述的条件或事件)时”或“响应于检测到(所陈述的条件或事件)”。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线)或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如DVD)、或者半导体介质(例如固态硬盘)等。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，该流程可以由计算机程序来指令相关的硬件完成，该程序可存储于计算机可读取存储介质中，该程序在执行时，可包括如上述各方法实施例的流程。而前述的存储介质包括：ROM或随机存储记忆体RAM、磁碟或者光盘等各种可存储程序代码的介质。

Claims (17)

  1. 一种广告显示方法，其特征在于，所述方法包括：
    所述电子设备获取用户的第一个人数据；所述第一个人数据为所述用户的个人信息；
    所述电子设备根据所述第一个人数据构建个人知识图谱；所述个人知识图谱包括所述第一个人数据和所述第一个人数据产生的时间；
    所述电子设备从广告服务器获取第一广告内容的参数信息；所述参数信息包括所述第一广告内容的类型、所述第一广告内容的链接地址；所述第一广告内容是所述广告服务器根据群体数据在多个广告中筛选得到的；所述第一广告内容包括一个或多个广告；
    所述电子设备根据所述个人知识图谱从所述第一广告内容的参数信息中获取到第二广告内容的参数信息；
    所述电子设备根据所述第二广告内容的参数信息获取到所述第二广告内容；所述第二广告内容包括一个或多个广告；
    所述电子设备在显示屏上显示所述第二广告内容。
  2. 根据权利要求1所述的方法，其特征在于，所述电子设备根据所述第一个人数据构建个人知识图谱，具体包括：
    所述电子设备从所述第一个人数据中获取到第二个人数据；所述第二个人数据包括关系知识、事件知识、实体知识；
    所述电子设备将所述关系知识、所述事件知识、所述实体知识按照预定结构存储；
    所述电子设备根据预定结构的关系知识、预定结构的事件知识、预定结构的实体知识构建所述用户的个人知识图谱。
  3. 根据权利要求1所述的方法，其特征在于，所述电子设备每隔固定周期获取所述用户的第一个人数据。
  4. 根据权利要求1所述的方法，其特征在于，所述第一广告内容是一下任意一种或几种：图片、视频、文字、音频。
  5. 根据权利要求2所述的方法，其特征在于，所述预定结构为五元组结构；
    所述电子设备将所述关系知识按照预定结构存储，具体包括：
    所述电子设备将所述关系知识按照第一五元组结构进行存储；所述第一五元组结构为“第一实体-关系-第二实体-第一时间点-第一时间区间”；所述关系表征所述第一实体与所述第二实体的关系，所述第一时间点为所述第一实体与所述第二实体建立所述关系的时间，所述第一时间区间为所述第一时间点到当前时间点的间隔时间。
  6. 根据权利要求2所述的方法，其特征在于，所述预定结构为五元组结构；
    所述电子设备将所述事件知识按照预定结构存储，具体包括：
    所述电子设备将所述事件知识按照第二五元组结构进行存储；所述第二五元组结构为“事件-论元-逻辑关系-第二时间点-第二时间区间”；所述论元为所述事件的发生动作，所述逻辑关系表征所述事件与所述论元的关系，所述第二时间点为所述事件发生的时间，所述第二时间区间为所述第二时间点到所述当前时间点的间隔时间。
  7. 根据权利要求2所述的方法，其特征在于，所述预定结构为五元组结构；
    所述电子设备将所述实体知识按照预定结构存储，具体包括：
    所述电子设备将所述实体知识按照第三五元组结构进行存储；所述第三五元组结构为“第三实体：第三时间点-第一关联权重-第四实体-第二关联权重-第五实体”；所述第三时间点为所述第三实体的发生时间，所述第一关联权重为所述第三实体与所述第四实体的关联程度，所述第二关联权重为所述第四实体与所述第五实体的关联程度。
  8. 根据权利要求5-7任一项所述的方法，其特征在于，所述方法包括：
    所述电子设备删除所述个人知识图谱中所述第一时间区间大于第一阈值的所述关系知识；
    和/或，
    所述电子设备删除所述个人知识图谱中所述第二时间区间大于所述第一阈值的所述事件知识；
    和/或，
    所述电子设备根据所述第三时间点确定出所述第三时间点到所述当前时间点的第三时间区间；
    所述电子设备删除所述个人知识图谱中所述第三时间区间大于所述第一阈值的所述实体知识。
  9. 根据权利要求1所述的方法，其特征在于，在所述电子设备根据所述用户的第一个人数据构建个人知识图谱之后，所述方法还包括：
    所述电子设备获取到用户历史行为和所述电子设备显示的历史广告信息；
    所述电子设备将所述历史广告信息、所述个人知识图谱作为所述重排序模型的输入，所述重排序模型输出第一结果；
    所述电子设备将所述第一结果与所述用户历史行为比较，并修改所述重排序模型的参数，直至所述重排序模型输出的所述第一结果在预设范围内，得到所述第一模型；
    所述电子设备根据所述个人知识图谱从所述第一广告内容的参数信息中获取到第二广告内容的参数信息，具体包括：
    所述电子设备根据所述第一模型从所述第一广告内容的参数信息中获取到第二广告内容的参数信息。
  10. 根据权利要求9所述的方法，其特征在于，所述电子设备根据所述第一模型从所述第一广告内容的参数信息中获取到第二广告内容的参数信息，具体包括：
    所述电子设备根据所述第一模型将所述第一广告内容的类型按照所述用户的喜爱度预测值从高到低进行排序，得到所述第二广告内容的参数信息；
    或者，
    所述电子设备根据所述第一模型将所述第一广告内容的类型按照所述用户的喜爱度预测值从高到低进行排序，并获取所述用户的喜爱度预测值高于第一阈值的广告的类型，得到所述第二广告内容的参数信息。
  11. 根据权利要求2所述的方法，其特征在于，在所述电子设备从所述第一个人数据中获 取到第二个人数据之前，所述方法还包括：
    所述电子设备将所述第一个人数据转化为文本信息；
    所述电子设备对所述文本信息进行断句、分词和词性标注；
    所述电子设备从所述第一个人数据中获取到第二个人数据，具体包括：
    所述电子设备获取所述文本信息中属于预设词性的词语。
  12. 根据权利要求2所述的方法，其特征在于，在所述电子设备获取所述文本信息中属于预设词性的词语之后，所述方法还包括：
    所述电子设备获取所述文本信息中出现次数为一次的词语；
    若所述文本信息中有出现两个及两个以上相同的词语，所述电子设备获取所述文本信息中有出现所述两个及两个以上相同的词语中的一个词语，得到所述第二个人数据。
  13. 根据权利要求1所述的方法，其特征在于，所述用户的个人信息包括以下一项或多项：性别、年龄、性格、爱好、人际关系、收入、通讯录信息、通话记录、短信、备忘录信息、居住的地址、所述居住的地址的天气情况。
  14. 根据权利要求2所述的方法，其特征在于，所述电子设备在显示屏的广告显示区域显示所述第二广告内容，具体包括：
    所述电子设备按照第二广告内容中所述用户的喜爱度预测值从高到低播放所述第二广告内容中的一个或多个广告；
    或者，
    所述电子设备显示所述第二广告内容中所述用户的喜爱度预测值最高的广告；
    或者，
    所述电子设备按照所述第二广告内容中所述用户的喜爱度预测值从高到低播放所述第二广告内容中的一个或多个广告，并屏蔽掉所述第二广告内容中所述电子设备在第一时间段内播放过的一个或多个广告。
  15. 根据权利要求1-14任一项所述的方法，其特征在于，在所述电子设备在显示屏的广告显示区域显示所述第二广告内容之后，所述方法还包括：
    所述电子设备获取所述用户对所述第二广告内容的观看数据；所述观看数据包括所述用户观看了所述第二广告内容中的一个或多个广告的广告类型和所述用户关闭了所述第二广告内容中的一个或多个广告的广告类型；
    所述电子设备根据所述观看数据更新所述第一模型。
  16. 一种电子设备，其特征在于，所述电子设备包括一个或多个处理器、一个或多个存储器、显示屏；所述一个或多个存储器、所述显示屏与所述一个或多个处理器耦合，所述一个或多个存储器用于存储计算机程序代码，所述计算机程序代码包括计算机指令，一个或多个处理器调用所述计算机指令以使得所述装置执行如权利要求1至15任一项所述的方法。
  17. 一种计算机可读存储介质，包括指令，其特征在于，当所述指令在电子设备上运行时，使得所述电子设备执行如权利要求1至15任一项所述的方法。
PCT/CN2021/117991 2020-09-19 2021-09-13 广告显示方法及电子设备 Ceased WO2022057764A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP21868587.3A EP4202730A4 (en) 2020-09-19 2021-09-13 ADVERTISING DISPLAY METHOD AND ELECTRONIC DEVICE
US18/027,040 US12412195B2 (en) 2020-09-19 2021-09-13 Advertisement display method and electronic device
US19/302,163 US20250371583A1 (en) 2020-09-19 2025-08-18 Advertisement display method and electronic device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010990422.3A CN114255056B (zh) 2020-09-19 2020-09-19 广告显示方法及电子设备
CN202010990422.3 2020-09-19

Related Child Applications (2)

Application Number Title Priority Date Filing Date
US18/027,040 A-371-Of-International US12412195B2 (en) 2020-09-19 2021-09-13 Advertisement display method and electronic device
US19/302,163 Continuation US20250371583A1 (en) 2020-09-19 2025-08-18 Advertisement display method and electronic device

Publications (1)

Publication Number Publication Date
WO2022057764A1 true WO2022057764A1 (zh) 2022-03-24

Family

ID=80776457

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/117991 Ceased WO2022057764A1 (zh) 2020-09-19 2021-09-13 广告显示方法及电子设备

Country Status (4)

Country Link
US (2) US12412195B2 (zh)
EP (1) EP4202730A4 (zh)
CN (1) CN114255056B (zh)
WO (1) WO2022057764A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117688230A (zh) * 2022-08-26 2024-03-12 腾讯科技（深圳）有限公司 一种模型训练方法、模型应用方法和相关装置

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112417117B (zh) * 2020-11-18 2024-07-23 腾讯科技（深圳）有限公司 会话消息生成方法、装置及设备
US12236453B2 (en) * 2021-12-03 2025-02-25 Jpmorgan Chase Bank, N.A. Systems and methods for providing offer interface platforms
CN114840117A (zh) * 2022-05-10 2022-08-02 北京字跳网络技术有限公司 信息输入页面的元素控制方法、装置、设备、介质
US12149764B2 (en) * 2022-10-13 2024-11-19 Adeia Guides Inc. Systems and methods for enhancing group media consumption sessions
US12225255B2 (en) 2022-10-13 2025-02-11 Adeia Guides Inc. Systems and methods for enhancing group media consumption sessions
US12562917B2 (en) * 2023-06-29 2026-02-24 International Business Machines Corporation Implementing personal knowledge graphs
CN121190126A (zh) * 2024-06-20 2025-12-23 华为技术有限公司 广告推荐方法及电子设备
CN118939875B (zh) * 2024-07-23 2025-01-28 上海市大数据中心 一种基于多模态数据的广告类案推荐系统及方法
CN120996875B (zh) * 2025-10-17 2026-01-23 西安诚优网络科技有限责任公司 一种基于强化学习的广告效果评估方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110066497A1 (en) * 2009-09-14 2011-03-17 Choicestream, Inc. Personalized advertising and recommendation
CN109447688A (zh) * 2018-09-27 2019-03-08 平安科技（深圳）有限公司 广告投放方法、装置及存储设备、计算机设备
CN110059249A (zh) * 2019-04-03 2019-07-26 华为技术有限公司 个性化推荐的方法、终端设备和系统
CN111582951A (zh) * 2020-05-28 2020-08-25 广州蓝深科技有限公司 一种用于云电子商务的广告投放系统和方法

Family Cites Families (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8359319B2 (en) * 2007-08-27 2013-01-22 Sudhir Pendse Tool for personalized search
US11017436B1 (en) * 2008-03-04 2021-05-25 Conviva Inc. Advertising engine
US20130268396A1 (en) * 2011-09-06 2013-10-10 Sony Mobile Communications Ab Method and system for providing personalized application recommendations
US9305307B2 (en) * 2013-07-15 2016-04-05 Google Inc. Selecting content associated with a collection of entities
US20150112818A1 (en) * 2013-10-22 2015-04-23 Google Inc. Content item selection criteria generation
US20170148068A1 (en) * 2014-05-09 2017-05-25 Google Inc. Entity based content distribution
WO2018209254A1 (en) * 2017-05-11 2018-11-15 Hubspot, Inc. Methods and systems for automated generation of personalized messages
US20160203137A1 (en) * 2014-12-17 2016-07-14 InSnap, Inc. Imputing knowledge graph attributes to digital multimedia based on image and video metadata
US20170024375A1 (en) * 2015-07-26 2017-01-26 Microsoft Technology Licensing, Llc Personal knowledge graph population from declarative user utterances
US11188830B2 (en) 2016-03-01 2021-11-30 Verizon Media Inc. Method and system for user profiling for content recommendation
CN107369098B (zh) * 2016-05-11 2021-10-26 华为技术有限公司 社交网络中数据的处理方法和装置
CN107944898A (zh) * 2016-10-13 2018-04-20 驰众信息技术（上海）有限公司 广告投放楼宇信息的自动发现与排序方法
US10489826B2 (en) * 2016-12-27 2019-11-26 Rovi Guides, Inc. Systems and methods for submitting user selected profile information to an advertiser
CN106997549A (zh) 2017-02-14 2017-08-01 火烈鸟网络(广州)股份有限公司 一种广告信息的推送方法及系统
JP6898892B2 (ja) * 2018-08-15 2021-07-07 Ｎｅｃパーソナルコンピュータ株式会社 電子機器及び広告表示方法並びに広告表示プログラム
KR102845942B1 (ko) * 2018-10-16 2025-08-12 삼성전자주식회사 지식 그래프에 기초하여 콘텐트를 제공하는 시스템 및 방법
CN109447713A (zh) * 2018-10-31 2019-03-08 国家电网公司 一种基于知识图谱的推荐方法及装置
CN111291191B (zh) 2018-12-07 2024-05-03 国家新闻出版广电总局广播科学研究院 一种广电知识图谱构建方法及装置
CN111353091B (zh) * 2018-12-24 2024-10-11 北京三星通信技术研究有限公司 信息处理方法、装置、电子设备及可读存储介质
CN109800300A (zh) * 2019-01-08 2019-05-24 广东小天才科技有限公司 一种学习内容推荐方法及系统
KR102079289B1 (ko) * 2019-04-23 2020-04-07 주식회사 비닛 와인 추천 시스템 및 방법
CN110458641B (zh) * 2019-06-28 2022-02-25 苏宁云计算有限公司 一种电商推荐方法及系统
US11379883B2 (en) * 2019-08-09 2022-07-05 SOCI, Inc. Systems, devices, and methods for dynamically generating, distributing, and managing online communications
CN111221983B (zh) * 2020-01-15 2023-08-04 北京百度网讯科技有限公司 时序知识图谱生成方法、装置、设备和介质
CN111768231B (zh) * 2020-06-24 2024-04-09 中国工商银行股份有限公司 产品信息推荐方法及装置
CN111625659B (zh) * 2020-08-03 2020-11-13 腾讯科技（深圳）有限公司 知识图谱处理方法、装置、服务器及存储介质
CN112102002A (zh) * 2020-09-16 2020-12-18 珠海格力电器股份有限公司 基于知识图谱的商品推荐方法、装置、计算机和存储介质
CN112287125B (zh) * 2020-11-23 2025-04-04 深圳季连科技有限公司 一种知识图谱的构建方法与系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110066497A1 (en) * 2009-09-14 2011-03-17 Choicestream, Inc. Personalized advertising and recommendation
CN109447688A (zh) * 2018-09-27 2019-03-08 平安科技（深圳）有限公司 广告投放方法、装置及存储设备、计算机设备
CN110059249A (zh) * 2019-04-03 2019-07-26 华为技术有限公司 个性化推荐的方法、终端设备和系统
CN111582951A (zh) * 2020-05-28 2020-08-25 广州蓝深科技有限公司 一种用于云电子商务的广告投放系统和方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4202730A4

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117688230A (zh) * 2022-08-26 2024-03-12 腾讯科技（深圳）有限公司 一种模型训练方法、模型应用方法和相关装置

Also Published As

Publication number Publication date
EP4202730A4 (en) 2024-01-24
CN114255056A (zh) 2022-03-29
US20250371583A1 (en) 2025-12-04
EP4202730A1 (en) 2023-06-28
US20230325884A1 (en) 2023-10-12
US12412195B2 (en) 2025-09-09
CN114255056B (zh) 2025-03-25

Similar Documents

Publication Publication Date Title
CN114255056B (zh) 广告显示方法及电子设备
CN109952572B (zh) 基于消息贴纸的建议响应
JP6419993B2 (ja) 関連コンテンツを先見的に特定し、タッチ感知デバイス上に表面化させるためのシステム及び方法
CN110110203B (zh) 资源信息推送方法及服务器、资源信息展示方法及终端
EP2959485B1 (en) Interaction with a portion of a content item through a virtual assistant
WO2021018154A1 (zh) 信息表示方法及装置
CN110458360B (zh) 热门资源的预测方法、装置、设备及存储介质
KR101917070B1 (ko) 이동 단말기, 서버, 시스템, 이동 단말기 및 서버의 제어 방법
WO2020238356A1 (zh) 界面显示方法、装置、终端及存储介质
WO2019140703A1 (zh) 一种用户画像的生成方法及装置
US20160179899A1 (en) Method of providing content and electronic apparatus performing the method
CN107305566B (zh) 一种为搜索信息匹配图片的方法及装置
CN107408120A (zh) 电子设备及其图像显示方法
CN104035995A (zh) 群标签生成方法及装置
CN118446750B (zh) 应用推荐方法、电子设备及云服务器
CN112000823B (zh) 功能入口显示方法、电子设备及计算机可读存储介质
WO2023125692A1 (zh) 一种服务推荐方法及相关装置
US20170249325A1 (en) Proactive favorite leisure interest identification for personalized experiences
CN117807123A (zh) 一种服务卡片推荐的方法和电子设备
CN108111603A (zh) 信息推荐方法、装置、终端设备及存储介质
WO2024174985A1 (zh) 广告文案生成方法、网络设备和计算机存储介质
CN115809362A (zh) 内容推荐方法及电子设备
KR101743999B1 (ko) 컨텐츠 검증 단말기, 시스템 및 그 것들의 제어방법
CN121210441A (zh) 一种记忆信息的管理方法和设备
CN114780828A (zh) 资源推荐方法、装置、计算机设备及介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21868587

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021868587

Country of ref document: EP

Effective date: 20230322

NENP Non-entry into the national phase

Ref country code: DE

WWG Wipo information: grant in national office

Ref document number: 18027040

Country of ref document: US