Presentation 1996/5/17
A Statistical Method for Merging Partial Parses
Naoya Arakawa, Toshiyuki Takezawa, Tsuyoshi Morimoto,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents a statistical method for analyzing naturally spoken sentences which are not so grammatical that conventional methods cannot parse. Conventional parsers based on phrase structure grammar often fail to parse disfluent utterances, and yield partial parses (or partial trees). When partial parses constitute a sentence (or are governed by a single predicative expression such as a verb), merging them into a single parse should serve for natural language understanding or machine translation. A statistical method was used for a robust merging process.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) spoken language processing / dependency structure / partial parses / robust parsing / stochastic language modeling
Paper # NCL96-8
Date of Issue

Conference Information
Committee NLC
Conference Date 1996/5/17(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) A Statistical Method for Merging Partial Parses
Sub Title (in English)
Keyword(1) spoken language processing
Keyword(2) dependency structure
Keyword(3) partial parses
Keyword(4) robust parsing
Keyword(5) stochastic language modeling
1st Author's Name Naoya Arakawa
1st Author's Affiliation ATR Interpreting Telecommunications Research Laboratories()
2nd Author's Name Toshiyuki Takezawa
2nd Author's Affiliation ATR Interpreting Telecommunications Research Laboratories
3rd Author's Name Tsuyoshi Morimoto
3rd Author's Affiliation ATR Interpreting Telecommunications Research Laboratories
Date 1996/5/17
Paper # NCL96-8
Volume (vol) vol.96
Number (no) 46
Page pp.pp.-
#Pages 8
Date of Issue