Method and apparatus for document filtering capable of efficiently extracting document matching to searcher's intention using learning data

Abstract

A document filtering device includes an information input / output unit, a search word extraction unit, a first sequence retrieval unit, a learning data unit, a classification parameter generation unit, a second sequence retrieval unit, and a classification unit. The information input / output unit inputs phrase information and outputs retrieval result information. The search word extraction unit extracts the search word from the phrase information. The first sequence retrieval unit retrieves documents having the search terms from the database, and outputs the first sequence retrieval results. The learning data unit prepares learning data from the first sequence retrieval result. A classification parameter generation unit generates classification parameters from the learning data. The second sequential retrieval unit retrieves documents having words corresponding to the classification parameters from the database. The classification unit extracts documents that match the searcher's intention and outputs the documents as second-order search results.

Description

[0001] This application claims priority from Japanese Patent Application Serial No. 2003-329206 filed in the Japan Patent Office on September 19, 2003, the entire contents of which are hereby incorporated by reference. technical field [0002] The present invention relates to a method and device for document filtering, in particular to a document filtering method and device capable of effectively extracting documents matching a searcher's intention from a document database by using learning data. Background technique [0003] How to effectively retrieve documents matching the searcher's intention from the database has become a problem. In order to solve the above-mentioned problems, traditional document retrieval technology uses a combination of keywords and logical operators to perform retrieval to obtain retrieval results, and subsequent retrieval uses a new combination of keywords and logical operators to refine the retrieval results. [0004] However, a searcher needs sp...

AI Technical Summary

Problems solved by technology

[0007] However, the retrieval results obtained by the above-mentioned conventional techniques may contain document data unnecessary for the searcher, and there is a disadvantage that it cannot clearly distinguish the data necessary for the searcher from the unnecessary data from unknown documents. necessary data

Images