Presentation 1999/7/23
Integrated Japanese Dependency Analysis using Rules and Corpus
Hiroyuki Takahasi, Masahiro Miyazaki,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) It is necessary to improve dependency analysis of Japanese subordinate clauses for precise analysis of long sentences. This paper proposes a integrated approach using rule-basted method and corpus-based statistical method. As for extraction of statistical information from corpus, we don't use any classification of clauses preventing risk of biased classification, but uses its appearance directly. We integrate rule-based method and corpus-based statistical method, selecting suitable method between two methods according as raliability of each result. Using this method, the experment shows that accuracy of the dependency relation between clauses is improved.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) dependency analysis / Japanese subordinate clause / statistical analysis / rule-based analysis
Paper # NLC99-16
Date of Issue

Conference Information
Committee NLC
Conference Date 1999/7/23(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) Integrated Japanese Dependency Analysis using Rules and Corpus
Sub Title (in English)
Keyword(1) dependency analysis
Keyword(2) Japanese subordinate clause
Keyword(3) statistical analysis
Keyword(4) rule-based analysis
1st Author's Name Hiroyuki Takahasi
1st Author's Affiliation The graduate school of Science and Technology, Niigata University()
2nd Author's Name Masahiro Miyazaki
2nd Author's Affiliation The graduate school of Science and Technology, Niigata University
Date 1999/7/23
Paper # NLC99-16
Volume (vol) vol.99
Number (no) 228
Page pp.pp.-
#Pages 6
Date of Issue