Presentation 2002/5/20
A QA Document Retrieval Method based on Phrase-Level Analysis of the Input Natura Language Query.
Motoyuki Itoh, Masateru Kubodera,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In the field of help-desk business, the need for an accurate QA document retrieval system is intensively increasing. Many QA document retrieval systems have been realized by simply applying conventional full-text search engines designed for general search purposes. But the output of those systems contains not a few "noise" documents less concerned with the content of input query, because they merely treat input query sentences as a set of keywords for search and don't consder with the intentional aspects or semantic content of the sentences. To reduce this noise-jamming problem, we are now developing a new search method that uses phrase-level semantic constraints derived from the input query to select relevant documents to the query. In this paper, we described the search method and the experimental system based on it.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Natural Language Information Retrieval / FAQ Retrieval / Document Retrieval / Natural Language Processing
Paper # TL2002-6
Date of Issue

Conference Information
Committee TL
Conference Date 2002/5/20(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 Thought and Language (TL)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A QA Document Retrieval Method based on Phrase-Level Analysis of the Input Natura Language Query.
Sub Title (in English)
Keyword(1) Natural Language Information Retrieval
Keyword(2) FAQ Retrieval
Keyword(3) Document Retrieval
Keyword(4) Natural Language Processing
1st Author's Name Motoyuki Itoh
1st Author's Affiliation CSK Corporation()
2nd Author's Name Masateru Kubodera
2nd Author's Affiliation CSK Corporation
Date 2002/5/20
Paper # TL2002-6
Volume (vol) vol.102
Number (no) 92
Page pp.pp.-
#Pages 6
Date of Issue