| 摘要 |
PROBLEM TO BE SOLVED: To solve the problem that automatic classification of all documents has low classification accuracy, and manual classification thereof needs much labor hour in hierarchical classification of documents according to the contents for finding an intended document, which is one of important subjects in the field of document management. SOLUTION: This device comprises a history storage part 103 storing the history of interactive search of a document by a user along classification layers, a registration candidate document storage part 104 storing a document the user desires to newly register when he/she cannot find the intended document; and a registration candidate document classification part 105 automatically classifying the registration candidate document and adding it as a new classification to the terminal of the existing classification layers or a higher-order classification. COPYRIGHT: (C)2007,JPO&INPIT
|
