Query Expansion with the Minimum Judgement(Text Mining II)

スポンサーリンク

概要

• 論文の詳細を見る

• Query expansion is one of feedback techniques in information retrieval, which needs a certain amount of relevance information that costs high in terms of human effort. In this paper we propose a method of query expansion which utilizes human help but with the minimum cost. Our purpose is to reduce users' cost when judging the relevancy of documents as much as possible using Transductive Learning. We describe this learning method is used to predict the relevancy of documents with no manual judgement based on only a fraction of true relevance information. We also show the role of the learning in our query expansion procedure. Compared with traditional query expansion methods, our method show the distinct effectiveness of query expansion, especially in the top 10 or 20 documents.

• 一般社団法人情報処理学会の論文
• 2004-12-04

著者

• YAMADA SEIJI

  National Institute of Informatics

• Okabe Masayuki

  Toyohashi University Of Technology

関連論文

• Relevance Feedback Document Retrieval using Non-Relevant Documents(Text Mining I)
• Action sloping for increasing awareness of robot's function (特集論文 HRI(Human-Robot Interaction))
• Relevance Feedback Document Retrieval using Non-Relevant Documents(Text Mining I)(Joint Workshop of Vietnamese Society of AI, SIGKBS-JSAI, ICS-IPSJ, and IEICE-SIGAI on Active Mining)
• Query Expansion with the Minimum Judgement(Text Mining II)
• Query Expansion with the Minimum Judgement(Text Mining I)(Joint Workshop of Vietnamese Society of AI, SIGKBS-JSAI, ICS-IPSJ, and IEICE-SIGAI on Active Mining)
• Document Retrieval through Time-dependent Events in the Web(Web Data Mining)
• Document Retrieval through Time-dependent Events in the Web(Web Data Mining)(Joint Workshop of Vietnamese Society of AI, SIGKBS-JSAI, ICS-IPSJ, and IEICE-SIGAI on Active Mining)

スポンサーリンク