Presentation 2012-08-31
Extracting and Alerting Several Informations from Time Depemdent Public Text Data
Tomoya TSUKAHARA, Katsutoshi TAKANASHI, Hideaki MIYAUCHI, Shun SATO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Time dependent documents (daily reports, papers/magazines, user generated web contents etc.) contain topics; some are sustained through several time-span and the others appear suddenly. We report a method to analyze both gradual transition of topics and upcoming topics. The time dependent clustering method weights not only keywords in existing clusters, but also timely keywords and encourages creating new clusters. Adopting the method to test data showed performance to visualize topic transition and discover new topics.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Document Clustering / Constrained Clustering / Temporal Information / Alert Information
Paper # NLC2012-19
Date of Issue

Conference Information
Committee NLC
Conference Date 2012/8/23(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) Extracting and Alerting Several Informations from Time Depemdent Public Text Data
Sub Title (in English)
Keyword(1) Document Clustering
Keyword(2) Constrained Clustering
Keyword(3) Temporal Information
Keyword(4) Alert Information
1st Author's Name Tomoya TSUKAHARA
1st Author's Affiliation Hitachi East Japan Solutions, Ltd.()
2nd Author's Name Katsutoshi TAKANASHI
2nd Author's Affiliation Hitachi East Japan Solutions, Ltd.
3rd Author's Name Hideaki MIYAUCHI
3rd Author's Affiliation Hitachi East Japan Solutions, Ltd.
4th Author's Name Shun SATO
4th Author's Affiliation Hitachi East Japan Solutions, Ltd.
Date 2012-08-31
Paper # NLC2012-19
Volume (vol) vol.112
Number (no) 196
Page pp.pp.-
#Pages 5
Date of Issue