Presentation 2013-07-23
マイクロブログ文書の選択による擬似適合フィードバック(一般講演,情報検索(2),ビッグデータを対象とした管理・情報検索・知識獲得及び一般)
Taiki MIYANISHI, Kazuhiro SEKI, Kuniaki UEHARA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Query expansion method using pseudo-relevance feedback is significantly effective for microblog search because the query expansion method can solve the vocabulary mismatch problem when searching short documents such as twitter messages (i.e. tweet) limited to 140 characters of content. The pseudo-relevance feedback approaches assume that the top ranked documents in the initial search results are relevant, which contain good topic-related words for relevance feedback. However, that assumption is often invalid because the initial search results may largely contain non-relevant documents and only a small share of the suggested expansion words are useful. Many others are either harmful or useless. To overcome the limitation of pseudo-relevance feedback for microblog search, we pro-pose novel query expansion method based on two-stage relevance feedback frameworks modeling search interests by manual tweet selection and integrating lexical and temporal evidence into its relevance model. Our experiments on a corpus of microblog data (Tweets2011 corpus) show that our two-stage relevance feedback approaches considerably improve search result relevance without decreasing retrieval over almost all topics.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Twitter / Microblog Search / Pseudo-Relevance Feedback / Query Expansion
Paper # DE2013-21
Date of Issue

Conference Information
Committee DE
Conference Date 2013/7/15(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)
Sub Title (in English)
Keyword(1) Twitter
Keyword(2) Microblog Search
Keyword(3) Pseudo-Relevance Feedback
Keyword(4) Query Expansion
1st Author's Name Taiki MIYANISHI
1st Author's Affiliation Graduate School of System Informatics Kobe University Japan.()
2nd Author's Name Kazuhiro SEKI
2nd Author's Affiliation Graduate School of System Informatics Kobe University Japan.
3rd Author's Name Kuniaki UEHARA
3rd Author's Affiliation Graduate School of System Informatics Kobe University Japan.
Date 2013-07-23
Paper # DE2013-21
Volume (vol) vol.113
Number (no) 150
Page pp.pp.-
#Pages 6
Date of Issue