Presentation 2000/7/19
Semantic Relevance Feedback for Web Document Retrieval
YOUICHI HIRATA, TAKESHI MATSUKURA, KEISHI TAJIMA, KATSUMI TANAKA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In conventional relevance feedback for Web query systems, the user evaluates some smaple pages, and then based on that evaluation, the original query is revised, or the query results are reorganized. In that evaluation, sample pages are classified as good or bad. This approach is effective to find pages similar to the pages evaluated as good. In some cases, however, the users want to specify their requirements more specifically, such as, "I want pages discussing this topic in more detail." In the usual relevance feedback, the users cannot express such requirements in the evaluation. In this paper, we propose a framework where the users can express such requirements, which we call semantic relevance feedback. In our framework, the users specify the difference between a sample page and pages they want. We estimate the difference between pages based on the amount of information and the extent of the topic in each page. We measure the former by the number of word occurrences, and measure the latter by the ratio of cooccurrence of words in pages.
Keyword(in Japanese) (See Japanese page)
Keyword(in English)
Paper # DE2000-40
Date of Issue

Conference Information
Committee DE
Conference Date 2000/7/19(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Data Engineering (DE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Semantic Relevance Feedback for Web Document Retrieval
Sub Title (in English)
Keyword(1)
1st Author's Name YOUICHI HIRATA
1st Author's Affiliation Graduate School of Science and Technology, Kobe University()
2nd Author's Name TAKESHI MATSUKURA
2nd Author's Affiliation Graduate School of Science and Technology, Kobe University
3rd Author's Name KEISHI TAJIMA
3rd Author's Affiliation Department of Computer and Systems engineering, Kobe University
4th Author's Name KATSUMI TANAKA
4th Author's Affiliation Graduate School of Science and Technology, Kobe University
Date 2000/7/19
Paper # DE2000-40
Volume (vol) vol.100
Number (no) 226
Page pp.pp.-
#Pages 8
Date of Issue