Presentation 2005/7/15
Latest Topic Words Detection from Chronological News Stream
Yoshihide SATO, Harumi KAWASHIMA, Tsutomu SASAKI, Masahiro OKU,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a method to detect 'Latest Topic Words' from incoming news articles to understand topical overview in them. Each word is representation of latest topical incident. Two aspects of topic in news stream have to be considered. One is accession of articles which play up same event, and the other is a continuation of follow up articles. At first we apply clastering method to news collection, evaluate significance of each article, and then extract a few latest topic words from each cluster using the significance. We estimated our method by experiment using 2164 news articles with time stamp and made sure the effect of it.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) topic / keywords / clustering / news / similarity / freshness
Paper # NLC2005-1
Date of Issue

Conference Information
Committee NLC
Conference Date 2005/7/15(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 Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Latest Topic Words Detection from Chronological News Stream
Sub Title (in English)
Keyword(1) topic
Keyword(2) keywords
Keyword(3) clustering
Keyword(4) news
Keyword(5) similarity
Keyword(6) freshness
1st Author's Name Yoshihide SATO
1st Author's Affiliation NTT Cyber-Solutions Laboratories, NTT Corporation()
2nd Author's Name Harumi KAWASHIMA
2nd Author's Affiliation NTT Cyber-Solutions Laboratories, NTT Corporation
3rd Author's Name Tsutomu SASAKI
3rd Author's Affiliation NTT Cyber-Solutions Laboratories, NTT Corporation
4th Author's Name Masahiro OKU
4th Author's Affiliation NTT Cyber-Solutions Laboratories, NTT Corporation
Date 2005/7/15
Paper # NLC2005-1
Volume (vol) vol.105
Number (no) 203
Page pp.pp.-
#Pages 6
Date of Issue