Being the time serial document intensive manner which also the axial keyword which is the central keyword which

摘要

PROBLEM TO BE SOLVED: To collect documents handling the same subject matter with a document having a plain style, such as a diary and one's thoughts, described in an arbitrary style as a target.;SOLUTION: A time-series document grouping device analyzes an input document, acquires a general word that is a word in the document and the appearance position information of the general word, acquires a keyword that becomes the candidate of a main keyword and the appearance position information of the keyword in the input document, totals the keyword, selects a frequent keyword, divides a document including the selected keyword into a subset, determines whether a document in the subset describes the same topic, based on the result of the totalled general word appearing within a predetermined range around the keyword in the document in the subset, integrates the determination result for each subset, and finally determines whether the keyword is the center of the topic.;COPYRIGHT: (C)2007,JPO&INPIT