Presentation 2004/2/12
Japanese segmentation with only statistical property in a document (Thought and Language)
Hisashi KAMIMURA, Kunio OISHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Machine translation became practical with improvement in the speed of a computer. Furthermore, the accuracy of the natural language processing by the computer is improving in the increase in accumulation of the word knowledge accompanying the increase in a storage capacity. However, if word knowledge is not always updated, it will become impossible to correspond to a new word, and the accuracy of natural language processing will become low. Moreover, it is actual that people are raising accuracy as for word knowledge or the corpus. In this paper, the method of performing Japanese segmentation only using the statistical character in a document is proposed without using word knowledge.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Japanese segmentation / PPM / graph / variable length n-gram
Paper # TL2003-33,PRMU2003-219
Date of Issue

Conference Information
Committee TL
Conference Date 2004/2/12(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Vice Chair

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) Japanese segmentation with only statistical property in a document (Thought and Language)
Sub Title (in English)
Keyword(1) Japanese segmentation
Keyword(2) PPM
Keyword(3) graph
Keyword(4) variable length n-gram
1st Author's Name Hisashi KAMIMURA
1st Author's Affiliation Graduate School of System Electronics, Tokyo University of Technology()
2nd Author's Name Kunio OISHI
2nd Author's Affiliation Electronic departments, Tokyo University of Technology
Date 2004/2/12
Paper # TL2003-33,PRMU2003-219
Volume (vol) vol.103
Number (no) 656
Page pp.pp.-
#Pages 6
Date of Issue