To clarify such a relationship between subjects that a new subject (unsolved subject) is generated as a result of solving a certain subject.
A subject relationship data base construction part 101 extracts a subject word, quotation relationship of a patent document, content explanation of a quotation destination document, and the unsolved subject from specification text data of the patent document, and associates them with the patent document and accumulates them in a subject relationship data base 114. A subject relationship data base utilization part 106 utilizes the subject relationship data base 114 to narrow down the patent document from the subject word of the patent document. A content explanation clustering part 109 performs clustering about the content explanation of the narrowed down patent document, and an association part 110 associates the unsolved subject with the content explanation of the most significant cluster. By the association, the relationship between the subject of the patent document and the unsolved subject generated from the subject is clarified. A display unit 112 visually displays the association results.
COPYRIGHT: (C)2011,JPO&INPIT
JP2008282222A | ||||
JP2005326922A |
小川知也、外1名,技術文書引用情報の可視化,情報処理学会研究報告,社団法人情報処理学会,2003年 3月28日,Vol.2003、No.37,pp.67~74
virtue Tamio Ei