Document Structure Analysis for the NTCIR-5 Patent Retrieval Task
Atsushi Fujii, Tetsuya Ishikawa · 2005
This paper describes our system participated in the Document and Passage Retrieval Subtasks at the NTCIR-5 Patent Retrieval Task. The purpose of these subtasks was the invalidity search, in which a patent application including a target claim is used to search documents that can invalidate the demand in the claim. Our system is characterized by the structure analysis for both target claim and entire application. The tar-get claim is segmented into components, each of which is used to produce an initial query. The structure of the application is used to enhance each query. The candi-dates of relevant documents are retrieved and ranked on a component-by-component basis. The final doc-ument list is obtained by integrating these document lists. All passages in each document are ranked ac-cording to the relevance to the target claim. We show the effectiveness of our system experimentally.