JPS58203565A - Inquiry response method in terminal system - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a terminal automatic response system for TSSi, which is a terminal automatic response system that applies a reasoning engine suitable for professional OA terminal systems.
In an application program that uses the +i end, processing can be executed using an interactive response method. However, there is a drawback in that the make-time response belonging to the user's personal one-length pattern must be input every time the TSS terminal is used.

The purpose of this explanation is to learn the interaction between the user and the T"S SAM terminal, and use the accumulated production rules and frames to automatically calculate the timing that can be easily inferred. It is an object of the present invention to provide a method for sending a response to a medium-side computer and reducing the troublesome input by a terminal user.

Similar patterns often occur in the user's responses to the timings output to the TSS 4.

This device has a No. 16 shunt device between the communication line interface part of the terminal and the input/output device handled by the user, and uses the shunted signals to learn the individual usage patterns of the user. It is considered to be collected and accumulated as a percentage. Then, when the accumulated response pattern occurs while the user is using the TSS terminal, it will be automatically interrupted without bothering the user, based on the inference by the inference engine of knowledge engineering. Tue 111. respond to the computer;

Is the following an implementation of the present invention? lJ will be explained. First, FIG. 1 shows the overall structure of the present invention. The program in the central computer 13 communicates with the TSS 4 terminal device 10 to make an interim response to the user and executes the process. The 'rss terminal communication line interface unit 9 sends and receives signals to and from the central computer 13. The display output section 5 displays the message from the medium-heat side computer 13 and the message from the recommendation@billion section 1 to the user via the display information section 7.

The keyboard input section 6 accepts inputs from k users.

The input information section 8 outputs the input from the -board input section 6 and the message from the inference engine section 1 to the central computer 13 via the TSS terminal communication line interface section 9. In this example, the display information section 7 and the input information section 8 constitute a signal shunt 10. The audio output unit 2 is an 'i'ss
This is conveniently used to alert the user, who has been away from the terminal 10 for a while, that a response is required.

The control input unit 3 is used to start the task of initializing the mode to the inference engine 1 and changing the settings of the intelligent response by the inference engine 1 to the central computer.

FIG. 2 shows the internal configuration of the inference engine section 1. (15) stores the control procedures for knowledge selection control processing and knowledge interpretation processing of the promoting M1 engine. The RAM 18 stores knowledge of the user's terminal response pattern to the central computer 13, which is manually stored from an auxiliary storage device 1t4 such as a flexible disk. The auxiliary dCtJi device control device 19 receives instructions from the microprocessor 14 and performs control. The display information interface 20 interfaces with the display information section 7. The input information section interface 21 interfaces with the input information section 7. The control information input section interface 16 performs an interface with the control input section. The audio output section interface 17 performs an interface with the audio output m2. The microprocessor 14 controls the entire reasoning engine.

FIG. 3 shows the task structure of the control procedure of the inference engine. The inference 9 control procedure 22 consists of a control input processing task 23 and an inference processing task 26. The above control input processing task 23
is inference) 1. It consists of an initial setting processing task 24 for the W function (such as how much the response based on the inference test is to be automatically performed), a task 25 for processing midway changes in the initial setting process, and η. The inference processing task 26 includes a knowledge selection system (goods) processing task 27 that controls the guessing process, and a knowledge selection system (goods) processing task 27 that interprets and executes the accumulated response patterns of individuals to the central computer. The knowledge interpretation processing task 28 consists of a knowledge interpretation processing task 28 that determines the production rule (determining which response pattern to adopt) among the response patterns of k users. A production rule interpretation processing task 29 executes interpretation, and a frame-type knowledge interpretation processing task 30 executes interpretation of frame-type knowledge (knowledge that instructs individual processing contents of adopted response patterns).
It consists of

Next, the operation of the system of the present invention will be explained. FIG. 4 shows a flowchart of the control procedure of the inference engine. The output from the central computer is monitored31), and the symbol string pattern matching is performed on the monitor information with the response patterns of individual users (for example, Fig. 5) stored in the form of production rules. It is determined which type of response pattern is occurring (32). The response to the frame detail # corresponding to the selected response type and the corresponding timing (an example is shown in Fig. 6) is extracted and displayed on the display output section 5, and if the user understands it, (determined by return signal),
Run to the central computer (34). Next, medium heat S
The i-millimeter computer and others are all monitored 35),
It is determined whether this monitor content continues to correspond to the same frame or not (36). If it is the same frame, proceed to 34; if it is frame #4, proceed to 32.

According to this embodiment, by using an inexpensive microprocessor, a professional who is unfamiliar with computers but who is familiar with business can use an 'I'S 8 terminal once again, and the user can use it for the purpose of providing a troublesome response. I can do it.

TSS 4 for professional OA is aimed at users who are business experts but are amateurs when it comes to computers, so it is praised for its convenience and ease of use.

According to the present invention, the part containing the items that the user of the TSS terminal must input for the actual information issued by the Chuo-Kucombutaka is processed by pattern matching using an inference engine. Since the past response patterns of individuals can be searched, the Ministry of History can be freed from the hassle of inputting similar responses every time from a terminal.

[Brief explanation of drawings]

Figure 1 is an overall configuration diagram of the present invention's Kazuo Oboroshi 1, Figure 2 is a configuration diagram of a practical chestnut oil control system of the present invention, and Figure 3 is a diagram of the procedure of the present invention. Task (configuration diagram, Figure 4 shows 7
+i control move: Aim flow diagram, Figure 6 is a diagram showing an example of a production rule, and Figure 6 is a diagram showing frame tally. 1... Reasoning engine section, 7... Table information section, 8... Input information 1st port'''A 2 port 3rd port 24 VJ4I ¥1 Figure 5 Figure 6

Claims (1)

[Claims] By providing a 1dt Em outside the signal at the input/output part with the terminal user in the TSS terminal device, and connecting it to the inference @kan which performs the logic using symbol string pattern matching, it is possible to The inference engine is operated using the knowledge (production rules and frames) of query response patterns described and accumulated using Seki's description style, and the central
] An automatic terminal response method that applies inference and sends a response to a computer.