DOCUMENT ANALYSIS APPARATUS AND PROGRAM

• US 20150199427A1
• Filed: 03/26/2015
• Published: 07/16/2015
• Est. Priority Date: 09/26/2012
• Status: Abandoned Application

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation), Toshiba Solutions Corporation (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation), Toshiba Solutions Corporation (Toshiba Corporation)
• Inventors: MIYABE, Yasunari, GOTO, Kazuyuki, IWASAKI, Hideki, ISOBE, Shozo, MATSUMOTO, Shigeru
• Application Number:
  US14/669,721
• Publication Number:
  US 20150199427A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/35 Clustering; Classification

Abstract

A document analysis apparatus according to an embodiment an acquisition unit acquires a plurality of words by analyzing a text included in each of a plurality of documents stored in a document storage unit. A first determination unit determines, for each of the acquired words, the presence/absence of a correlation between the word and at least two attributes designated by a user out of a plurality of attributes of the plurality of documents stored in the document storage unit. A second determination unit determines whether a determination result by the first determination unit matches a pattern designated by the user out of a plurality of patterns stored in a pattern storage unit. A presentation unit presents a word whose determination result by the first determination unit is determined to match the pattern designated by the user.

Claims

1. A document analysis apparatus comprising:

a document storage unit which stores a plurality of documents each of which includes a text formed from a plurality of words, has a plurality of attributes, and includes attribute values of the attributes;

a pattern storage unit which stores a plurality of patterns each representing presence/absence of a correlation between a word and each of at least two attributes out of the plurality of attributes;

an acquisition unit which acquires a plurality of words by analyzing the text included in each of the plurality of documents stored in the document storage unit;

a first determination unit which determines, for each of the acquired words, the presence/absence of the correlation between the word and at least two attributes designated by a user out of the plurality of attributes of the plurality of documents stored in the document storage unit;

a second determination unit which determines whether a determination result by the first determination unit matches a pattern designated by the user out of the plurality of patterns stored in the pattern storage unit; and

a presentation unit which presents a word whose determination result by the first determination unit is determined to match the pattern designated by the user.

2. The document analysis apparatus according to claim 1, further comprising:

a first calculation unit which calculates, for each word whose determination result is determined to match the pattern designated by the user, a degree of feature based on an appearance frequency of the word in the plurality of documents stored in the document storage unit; and

a second calculation unit which calculates, for each word whose determination result is determined to match the pattern designated by the user, a degree of association based on cooccurrence of the word in the plurality of documents stored in the document storage unit and a word other than the word, whose determination result by the first determination unit is determined to match the pattern designated by the user,

wherein the presentation unit presents the word whose determination result by the first determination unit is determined to match the pattern designated by the user, based on the degree of feature and the degree of association calculated for each word.

3. The document analysis apparatus according to claim 2, wherein the second calculation unit calculates, for each word whose determination result by the first determination unit is determined to match the pattern designated by the user, the degree of association based on cooccurrence of the word and a word whose cooccurrence frequency with the word is statistically significant.

4. The document analysis apparatus according to claim 1, further comprising a category generation unit,

wherein the at least two attributes designated by the user include a first attribute and a second attribute,

the category generation unit generates a first category into which the plurality of documents are classified based on an attribute value of the first attribute included in the plurality of documents, and generates a second category into which the plurality of documents are classified based on the attribute value of the second attribute included in the plurality of documents, and

the presentation unit further presents a cross tabulation result including the number of documents classified into both the first category and the second category, which are generated.

5. The document analysis apparatus according to claim 4, when the presented word is designated by the user, wherein the presentation unit presents the cross tabulation result including the number of documents classified into both the first category and the second category, which are generated, out of the documents including the word.

6. A program stored in a non-transitory computer-readable storage medium, the program being executed by a computer of a document analysis apparatus including a document storage unit which stores a plurality of documents each of which includes a text formed from a plurality of words, has a plurality of attributes, and includes attribute values of the attributes, and a pattern storage unit which stores a plurality of patterns each representing presence/absence of a correlation between a word and each of at least two attributes out of the plurality of attributes, the program causing the computer to execute an analysis method, the analysis method comprising:

acquiring a plurality of words by analyzing the text included in each of the plurality of documents stored in the document storage unit;

determining, for each of the acquired words, the presence/absence of the correlation between the word and at least two attributes designated by a user out of the plurality of attributes of the plurality of documents stored in the document storage unit;

determining whether a determination result matches a pattern designated by the user out of the plurality of patterns stored in the pattern storage unit; and

presenting a word whose determination result is determined to match the pattern designated by the user.