AUTOMATIC DOCUMENT CLASSIFICATION SYSTEM, AUTOMATIC DOCUMENT CLASSIFICATION METHOD, AND COMPUTER-READABLE RECORDING MEDIUM WITH AUTOMATIC DOCUMENT CLASSIFICATION PROGRAM RECORDED THEREON

摘要

PROBLEM TO BE SOLVED: To solve such problems that the classification category for documents to be classified has had to be prepared in advance and a large amount of documents for learning set for every this classification category have had to be prepared. SOLUTION: The automatic document classification system has a document structure analysis means that analyzes the structure of classification to be classified, the standard classification documents provided according to classification categories and the classified documents to be classified into classification categories, a similarity calculation means that calculates the similarity degree of the point of view between the clustered standard classification documents and a result of the analysis for the classified document, which is derived by conforming to a rule that specifies the point of view by clustering the standard documents for the classification based on the document structure analysis means, and a document classification means that decides the category of the documents to be classified based upon the similarity degree that the similarity calculation means has calculated.