JPH022458A - Similar document search device - 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 Field of the Invention The present invention relates to a similar document retrieval device used for processing pertinent information.

BACKGROUND OF THE INVENTION In recent years, there have been many opportunities to handle large amounts of databases.

An example of the above-mentioned conventional similar document search device will be described below with reference to the drawings.

FIG. 2 shows a configuration diagram of a conventional similar document search device. In FIG. 2, 21 is a keyword input section;
Reference numeral 2 denotes a keyword-added document memory for documents having keywords; 23 refers to a document extracting unit that extracts documents from the keyword-added document memory 22; and 26 refers to a keyword input unit 21 and keywords in the keyword-added document memory 22; A keyword similarity determination unit that compares and determines similarity; 2A is a similar document extraction instruction that instructs the document extraction unit 23 to extract a document with a high degree of similarity based on the keyword similarity obtained by the keyword similarity determination unit 26; Part, 28
is a display section that displays documents extracted by the document extraction section 23.

The operation of the similar document search device configured as described above will be described below.

First, a keyword is input into the keyword input section 21. Then, the keyword similarity determination section 26 compares the keyword input into the keyword input section 21 with the keyword in the keyword-attached document memory 22 to determine the degree of similarity. Next, the similar document extraction instruction section 27 instructs the document extraction section 23 to extract documents with high similarity based on the keyword similarity obtained by the keyword similarity determination section 26. The document extraction unit 23 retrieves the document instructed by the similar document extraction instruction unit 27 from the keyword-added document memory 22. The document extracted by the document extraction unit 23 is displayed on the display unit 28 (for example, “ASCII” 1
987.3 N1117 P, 105-120).

Problems to be Solved by the Invention However, with the above configuration, the documents that can be searched are limited to those that have keywords. There was a problem that it had to be done.

In view of the above-mentioned problems, the present invention provides a similar document retrieval device that can automatically extract keywords even for documents that do not have keywords by performing morphological analysis on the document to search for a desired document. It is something.

Means for Solving the Problems In order to solve the above problems, the similar document retrieval device of the present invention includes an input unit, a document memory, a document extraction unit that retrieves documents from the document memory, and a document extraction unit that retrieves documents from the document memory. A keyword extraction section that extracts keywords by performing morphological analysis of a document, a keyword memory that stores the keywords extracted by the keyword extraction section, and a keyword similarity measure that compares the keywords in the keyword memory with each other to determine the degree of similarity. a determination unit; a similar document extraction instruction unit that instructs a document extraction unit to extract documents with a high degree of similarity based on the keyword similarity obtained by the keyword similarity determination unit;
It is equipped with a display section.

Effect of the Invention The present invention provides a similar document retrieval device that is capable of automatically extracting keywords even for documents that do not have keywords by performing morphological analysis on the document and searching for a desired document. It becomes possible.

Example Below, regarding a similar document search device according to an example of the present invention,
This will be explained with reference to the drawings.

FIG. 1 shows a block diagram of a similar document search device according to a first embodiment of the present invention. In FIG. 1, 11 is an input unit, 12 is a document memory, 13 is a document extraction unit that extracts documents from the document memory 12, and 14 is a keyword that extracts keywords by performing morphological analysis on the document extracted by the document extraction unit 13. 16 is a keyword memory that stores the keyword extracted by the keyword extractor 14; 18 is a keyword similarity determination unit that compares the keywords in the keyword memory 16 with each other and determines the similarity; 17 is a keyword similarity determination unit; A similar sentence extraction instruction section instructs the document extraction section 13 to extract sentences with high similarity from the keyword similarity obtained in the keyword similarity section 16; 18 is a display for displaying documents extracted by the document extraction section 13; Department.

The operation of the similar document search device configured as described above will be described below with reference to FIG.

Here, it is assumed that the document stored in the document memory 12 may or may not have a keyword. First, information that allows a desired document to be searched for, such as a keyword or a document similar to the desired document, is input into the input unit 11.

Then, the document extracting section 13 takes out the document from the document memory 12, and the keyword extracting section 14 extracts keywords from the document taken out by the document extracting section 13 by performing morphological analysis or the like. Here, a method for extracting keywords by morphological analysis will be explained. When a sentence is morphologically analyzed, it is divided into words. Pick independent words from among the words and rearrange them in order of frequency. Then, select keywords in descending order of frequency. The keywords extracted by the keyword extraction unit 14 in this manner are stored in the keyword memory 16. Keyword similarity determination unit 16
compares the keywords in the keyword memory 16 with each other to determine similarity.

Then, the similar document extraction instruction unit 17 instructs the document extraction unit 13 to extract a document with a high degree of similarity based on the keyword similarity obtained by the keyword similarity determination unit 16. Therefore, the document extraction section 13 uses the similar document extraction instruction section 17.
A document with a high degree of similarity indicated by is retrieved from the document memory 12. The document extracted by the document extraction section 13 is then displayed on the display section 18.

As described above, according to this embodiment, there is an input section, a document memory, a document extraction section that extracts a document from the document memory, and a keyword extraction section that extracts keywords by performing morphological analysis on the document extracted by the document extraction section. a keyword memory that stores the keywords extracted by the keyword extraction section; a keyword similarity determination section that compares the keywords in the keyword memory with each other to determine the degree of similarity; and a keyword similarity determination section that stores the keywords extracted by the keyword extraction section. By providing a display unit and a similar document extraction instruction unit that instructs the document extraction unit to extract documents with a high degree of similarity based on the degree of similarity, it is possible to perform morphological analysis on documents that do not have keywords. This makes it possible to provide a similar document search device that can automatically extract keywords and search for a desired document.

Effects of the Invention As described above, the present invention includes an input section, a document memory, a document extraction section that extracts a document from the document memory, and a keyword extraction section that extracts keywords by performing morphological analysis on the document extracted by the sentence extraction section. an extraction unit, a keyword memory that stores the keywords extracted by the keyword extraction unit, a keyword similarity determination unit that compares the keywords in the keyword memory with each other and determines the degree of similarity, and a keyword similarity determination unit that stores the keywords extracted by the keyword extraction unit; By providing a display unit and a similar document extraction instruction unit that instructs the document extraction unit to extract documents with high similarity based on keyword similarity, it is possible to perform morphological analysis etc. on documents that do not have keywords. This will automatically extract keywords,
A similar document search device that can search for a desired document can be provided.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a similar document search device according to a first embodiment of the present invention, and FIG. 2 is a block diagram of a conventional similar document search device. 11...Input section, 12...Document memory, 13゜23...Document extraction section, 14...
・Keyword extraction unit, 15...Keyword memory, 16.26...Keyword similarity determination unit,
17.27... Similar document extraction instruction section, 18.2
8...Display section, 21...Keyword input section, 22...Document memory with keyword. Name of agent: Patent attorney Toshio Nakao and 1 other person 1st
figure

Claims (1)

[Claims] an input section, a document memory, a document extraction section that extracts a document from the document memory, a keyword extraction section that extracts keywords by performing morphological analysis on the document extracted by the document extraction section; a keyword memory for storing keywords that have been selected; a keyword similarity determination unit that compares the keywords in the keyword memory to determine the degree of similarity; and a keyword similarity determination unit that determines the degree of similarity from the keyword similarity obtained by the keyword similarity determination unit. A similar document retrieval device comprising: a similar document extraction instructing section for instructing the document extracting section to extract documents with high prices; and a display section.