Paper filtering method using features of coauthor research group, subject category, and terminology

Abstract

SUMMARYA paper filtering system that supports the effective collection of related technical papers is becoming important as technological progress accelerates. Two requirements for the paper filtering system are (1) reduction of the user workload in specifying the filtering conditions and (2) sufficient filtering accuracy. We propose a paper filtering method that meets both requirements simultaneously by focusing on the features of the coauthor research group, subject category, and terminology. The result of evaluation using 3600 domestic learned-society papers shows that the proposed method improved the mean average precision from 0.39 to 0.50, that is, by 0.11, compared with the conventional pseudo-relevance feedback method, thus improving its suitability for practical use.