Wise search engine based on LSI

Yang Jianxiong*, Junzo Watada

*この研究の対応する著者

研究成果: Conference contribution

抄録

The objective of this work is to provide, as a search engine, latent semantic indexing (LSI), which is a classical method to produce optimal approximations of a term-document matrix and has been used for textual information mining. The use of this technique is examining mine content which based web document, using keyword features of documents. Experimental results show that together with both textual and latent features LSI can extract the underlying semantic structure of web documents, thus improve the search engine performance significantly.

本文言語English
ホスト出版物のタイトルLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ページ126-136
ページ数11
5980 LNAI
DOI
出版ステータスPublished - 2010
外部発表はい
イベント6th International Workshop on Agents and Data Mining Interaction, ADMI 2010 - Toronto, ON
継続期間: 2010 5月 112010 5月 11

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
5980 LNAI
ISSN(印刷版)03029743
ISSN(電子版)16113349

Other

Other6th International Workshop on Agents and Data Mining Interaction, ADMI 2010
CityToronto, ON
Period10/5/1110/5/11

ASJC Scopus subject areas

  • コンピュータ サイエンス(全般)
  • 理論的コンピュータサイエンス

フィンガープリント

「Wise search engine based on LSI」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル