Presentation 1996/9/27
Evaluation of Algorithms for Unknown Word Acquisition
Chiaki KUBOMURA, Tomoko SAKURAI, Hiroyuki KAMEDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Advanced natural language processing systems must have abilities to acquire new words (henceforth, unknow words) automatically. From this point of view, we have implemented a prototype system with such abilities. In this paper, three acquisition algorithms for three classes of unknow words, which were adopted in the system, were experimentally examined with some amounts of linguistic material gathered from newspaper, books. magazines and lexicons. As a result, fundamental validity of the acquisition algoritms for unknown words was verified.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) knowledge acquisition / unknown word / acquisition algorithm / performance evaluation / natural language processing
Paper # TL96-6
Date of Issue

Conference Information
Committee TL
Conference Date 1996/9/27(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 Thought and Language (TL)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Evaluation of Algorithms for Unknown Word Acquisition
Sub Title (in English)
Keyword(1) knowledge acquisition
Keyword(2) unknown word
Keyword(3) acquisition algorithm
Keyword(4) performance evaluation
Keyword(5) natural language processing
1st Author's Name Chiaki KUBOMURA
1st Author's Affiliation Faculty of Engineering, Tokyo Engineering University()
2nd Author's Name Tomoko SAKURAI
2nd Author's Affiliation Department of Aesthetics & Health Science, Yamano College of Aesthetics
3rd Author's Name Hiroyuki KAMEDA
3rd Author's Affiliation Faculty of Engineering, Tokyo Engineering University
Date 1996/9/27
Paper # TL96-6
Volume (vol) vol.96
Number (no) 271
Page pp.pp.-
#Pages 10
Date of Issue