DOCUMENT ANALYSIS SYSTEM, DOCUMENT ANALYSIS METHOD, AND COMPUTER-READABLE RECORDING MEDIUM

摘要

A document analysis device (100) is provided with: a feature expression acquisition unit (10) which acquires feature expressions occurring during focusing on a analysis target document set; a document set acquisition unit (20) which, from an analysis parent set containing the analysis target document set, acquires a feature expression-containing document set consisting of documents in which the feature expressions occurred; a context determination unit (40) which, for each feature expression, specifies analysis / feature expression-containing documents, which are documents in the feature expression-containing document set that are analysis target documents, and specifies contexts of said feature expression which occur in two or more analysis / feature expression-containing documents; a context comparison determination unit (50) which specifies non-analysis / feature expression-containing documents, which are documents in the feature expression-containing document set that are not analysis target documents, and compares the contexts in which said feature expressions occur and the previously-specified contexts; and a feature value setting unit (60) which, on the basis of the comparison result, assigns feature values to the feature expressions.