Text search system for locating on the basis of keyword matching and keyword relationship matching

• US 5,297,039 A
• Filed: 01/27/1992
• Issued: 03/22/1994
• Est. Priority Date: 01/30/1991
• Status: Expired due to Fees

• Current Assignee: Mitsubishi Denki Kabushiki Kaisha (Mitsubishi Electric Corporation)
• Original Assignee: Mitsubishi Denki Kabushiki Kaisha (Mitsubishi Electric Corporation)
• Inventors: Koike, Kazuhiro, Taki, Hirokazu, Kanaegami, Atsushi, Ohgashi, Hitoshi
• Primary Examiner(s): Weinhardt, Robert A.
• Assistant Examiner(s): Shingala, Gita
• Application Number:
  US07/826,090
• Time in Patent Office: 785 Days
• Field of Search: 364/419, 395/63, 395/76
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: Y10S 706/934 Information retrieval or Information management
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

A text information extraction device extracts analysis networks from texts and stores them in a database. The analysis networks consist of lines each including elements and relations extracted from the texts. The analysis networks are complemented via synonym/near synonym/thesaurus process and via complementary template and the lines thereof are weighted via concept template. A text similarity matching device judges similarity of input and database analysis networks on the basis of agreements of words, word pairs, and lines. A text search system stores texts and complementary term lists prepared therefrom in respective databases. Queries are inputted in the form of analysis networks from which sets of keywords and relations are extracted. After searching the texts and complementary term lists stored in databases with respect to the keywords extracted from each input query, agreements of the sets of keywords and relations are determined.

Claims

1. A text search system comprising:

(a) a text database storing texts to be searched;

(b) a complementary term list database storing complementary term lists obtained from said texts via morphological analysis, each complementary term list including lexically divided words of one of said texts and terms related to said words;

(c) query analysis means for receiving an input query and outputting sets of keywords and relations between keywords contained in said input query;

(d) first search means for searching said text database with respect to said keywords obtained by said query analysis means, said first search means counting frequency of said keywords in respective texts stored in said text database and outputting texts in which said keywords occur at a frequency above a predetermined level;

(e) second search means for searching said complementary term list database with respect to said keywords obtained by said query analysis means, said second search means calculating similarity points of respective complementary term lists on the basis of weighted sum of frequency of said keywords in said respective complementary term lists, said second search means outputting complementary term lists having similarity points above another predetermined level;

(f) complementary term list generating means for generating complementary term lists of texts in response to said texts outputted from said first search means;

(g) third search means for searching among complementary term lists outputted from said second search means and said complementary term list generating means, for occurrences of said sets of keywords and relations outputted from said query analysis means.

2. A text search system as claimed in claim 1, further comprising:

(h) similarity judgment means for judging similarity of sets of keywords and relations by means of syntactical and semantic analysis, said similarity judgment means judging similarity between sets of keywords and relations outputted from said query analysis means and sets of keywords and relations which are found by said third search means in complementary term lists outputted from said second search means and said complementary term list generating means.

3. A text search system as claimed in claim 1, wherein said first search means includes means for adding synonyms/near synonyms to said keywords and counts frequency of said keywords and said synonyms/near synonyms thereof in respective texts stored in said text database.

4. A text search system comprising:

(a) keyword means for extracting keywords by means of morphological analysis from at least two texts to be matched;

(b) related word means for adding related words, including synonyms/near synonyms and thesaurus entries, to said keywords extracted from at least one of said texts, said related words and said keywords having associated therewith weighted degrees of similarity;

(c) judgment means for judging agreement of the keywords of one of said two texts with keywords of another one of said two texts; and

(d) evaluation means for evaluating similarity of keywords judged by said judgment means to agree with each other, said evaluation means including at least one of;

(d1) first evaluation means for changing weights of degrees of similarity in accordance with whether respective keywords judged by said judgment means to agree with each other are keywords extracted by said keyword means and keywords added by said related word means;

(d2) second evaluation means for changing weights of degrees of similarity when said agreements of keywords are partial agreements including forward, backward, and middle partial agreements, or

(d3) third evaluation means for changing weights of degrees of similarity in accordance with points of importance given beforehand to said keywords.

5. A text search system comprising:

a text database storing texts to be searched;

a knowledge database storing a word dictionary, a synonyms/related terms dictionary and a thesaurus;

means for performing morphological analysis so as to produce complementary term lists;

a complementary term list database for storing complimentary term lists generated from the text database and the knowledge database by the means for performing morphological analysis;

query analysis means for producing an output based on morphological analysis of a user defined input query; and

judgment means for receiving the output produced by the query analysis means, records from the text database and complementary term lists from the complementary term list database, the judgment means including means for judging similarity therebetween.

6. A text search system as recited in claim 5, wherein the query analysis means further comprises:

means for extracting keyword relationships based on the contents of the knowledge database.

7. A text search system as recited in claim 5 the query analysis means further comprising:

means for producing the output, wherein the output produced is a keyword complementary term list derived from the input query and the contents of the knowledge database.

8. A text search system as recited in claim 7, the judgment means further comprising:

means for judging similarity between the keyword complementary term list produced by the query analysis means and keywords in complementary term lists stored by the complementary term list database.