APPARATUS FOR ANALYZING SET OF DOCUMENTS, METHOD FOR ANALYZING SET OF DOCUMENTS, PROGRAM IMPLEMENTING THIS METHOD, AND RECORDING MEDIUM STORING THIS PROGRAM

摘要

PROBLEM TO BE SOLVED: To allow an apparatus for analyzing a set of documents to determine that even documents issued on different dates in terms of time are highly related to each other if their contents are highly correlated to each other.;SOLUTION: The apparatus for analyzing a set of documents includes a document set specifying part 10 for specifying a set of documents according to specification requirements; a content similarity evaluating part 20 for evaluating a similarity in content among the documents included in the set of documents; a time stamp similarity evaluating part 30 for evaluating a similarity in time among the documents included in the set of documents; a relationship extracting part 40 for extracting a relationship among the documents based on the similarities in content and in time; a centricity determining part 50 for calculating the centricity of the documents based on the relationship among the documents; an information analyzing part 100 for specifying a theme word contained in the set of documents, a set of documents related to the theme word, and the role of documents in the set of documents from the entire set of documents based on the relationship among the documents and the centricity of the individual documents obtained; and an information output part 110 for visualizing and outputting the specified set of documents.;COPYRIGHT: (C)2008,JPO&INPIT