Presentation 2011-10-10
A text segmentation method via continuous HMM
Yasuhiro TAJIMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A text segmentation method via continuous HMM is shown in this paper. In general, an HMM has output symbols of vocublary, i.e. one word will be output by one step. In contrast, our method uses an HMM whose output symbol is a vector of word counts. The vector corresponds to a sliding window on a text. We evaluate this HMM, then we confirmed that our method has higher performance than a general case HMM.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Text segmentation / n-gram / Hidden Markov Model
Paper # TL2011-30,NLC2011-27
Date of Issue

Conference Information
Committee NLC
Conference Date 2011/10/3(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) A text segmentation method via continuous HMM
Sub Title (in English)
Keyword(1) Text segmentation
Keyword(2) n-gram
Keyword(3) Hidden Markov Model
1st Author's Name Yasuhiro TAJIMA
1st Author's Affiliation Okayama Prefectural University, Department of systems engineering()
Date 2011-10-10
Paper # TL2011-30,NLC2011-27
Volume (vol) vol.111
Number (no) 228
Page pp.pp.-
#Pages 5
Date of Issue