表構造における意味的関係に基づくWWW検索精度の向上  [in Japanese] Improvement in precision of WWW Search Engines based on Semantic Relation in Table Structure  [in Japanese]

Access this Article

Search this Article

Author(s)

Abstract

Web検索エンジンに,ユーザが検索キーワードとして2つの語を入力した場合に,その2語が意味的関係を持って文書中に出現しているか否かを判定する,ランキング手法を提案する.キーワード間の意味的関係を表現する構造として,本研究では表構造を取り上げる.提案する手法に基づいて,既存検索エンジンの検索結果をフィルタリングするシステムを構築し,NTCIR-3WEBによるテストコレクションを用いて,本手法の精度を実験的に検証する.ln this paper,we propose a method to improve precision of WWW search engines. When Web search engines are used, the user inputs keywords. Web search engines accept a few keywords as a retrieval condition, and select candidate pages based on some statistics value such as TF/IDF and so on. We think that the keywords input by users have a semantic relation. We try to improve retrieval accuracy by checking whether two keywords have a semantic relation in the candidate pages or not. We deal with table as a structure to represent semantic relations between keywords. We construct a meta search engine which accept output of an ordinary search engine and select plausible ones by checking semantic relation. We use NTCIR-3WEB as a test collection,and we also show an experimental evaluation of our method.

In this paper, we propose a method to improve precision of WWW search engines. When Web search engines are used, the user inputs keywords. Web search engines accept a few keywords as a retrieval condition, and select candidate pages based on some statistics value such as TF/IDF and so on. We think that the keywords input by users have a semantic relation. We try to improve retrieval accuracy by checking whether two keywords have a semantic relation in the candidate pages or not. We deal with table as a structure to represent semantic relations between keywords. We construct a meta search engine which accept output of an ordinary search engine and select plausible ones by checking semantic relation. We use NTCIR-3WEB as a test collection, and we also show an experimental evaluation of our method.

Journal

  • IPSJ SIG Notes

    IPSJ SIG Notes 2006(58(2006-DD-055)), 5-11, 2006-05-26

    Information Processing Society of Japan (IPSJ)

References:  6

Cited by:  1

Codes

  • NII Article ID (NAID)
    110004822334
  • NII NACSIS-CAT ID (NCID)
    AN10539261
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    09196072
  • NDL Article ID
    7936103
  • NDL Source Classification
    ZM13(科学技術--科学技術一般--データ処理・計算機)
  • NDL Call No.
    Z14-1121
  • Data Source
    CJP  CJPref  NDL  NII-ELS  IR  IPSJ 
Page Top