Presentation 2007/7/17
Extracting Nouns that Constitute Templates by the Katz Model
Daisuke FUJIHARA, Akihiro TAKASE, Kyoji UMEMURA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A template is a fixed format of certain documents. We deal here with the problem of extraction words used in templates without knowing form of the templates. The Katz K mixture model is well known as a distribution model of keywords. In this model, basic assumption is that the conditional probabilities of repeats for a given word are determined by a decay factor. In this study, we analyze relations of a template and proper nouns which do not obey the Katz K mixture model. As a result, we have found that the Katz model is useful to detect nouns that consitute templates.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) the Katz K mixture model / statistical natural language processing / template / term frequency / proper noun
Paper # NLC2007-25
Date of Issue

Conference Information
Committee NLC
Conference Date 2007/7/17(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 Nouns that Constitute Templates by the Katz Model
Sub Title (in English)
Keyword(1) the Katz K mixture model
Keyword(2) statistical natural language processing
Keyword(3) template
Keyword(4) term frequency
Keyword(5) proper noun
1st Author's Name Daisuke FUJIHARA
1st Author's Affiliation Toyohashi University of Technology()
2nd Author's Name Akihiro TAKASE
2nd Author's Affiliation Toyohashi University of Technology
3rd Author's Name Kyoji UMEMURA
3rd Author's Affiliation Toyohashi University of Technology
Date 2007/7/17
Paper # NLC2007-25
Volume (vol) vol.107
Number (no) 158
Page pp.pp.-
#Pages 5
Date of Issue