Presentation 2011-10-10
A Refinement Method of Obtained Attributes for Undefined Concepts based on Temporal Alteration of Web
Noriyuki OKUMURA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper investigates a method of resolving the problem existing Auto Feedback method which makes existing Concept-base learn dynamically. Auto Feedback method uses search engines such as Google for obtaining associative words and their weights about undefined words in Concept-base. However, it happens some problems that the obtained associative words are different each searching time because the Internet varies from hour to hour. Temporary subjects/topics such as news, weblog, bulletin board and so on influence words' weights. Thus, in this paper, it is not made to learn by trial once, but two or more learning is performed continuously for a long period of time. By selecting associative words obtained in experiment period, we investigate a learning method for Concept-base dynamically. In addition, we verify the effectiveness of selecting words based on F-measure by evaluation experiment.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Concept-base / Auto Feedback / Obtaining Attributes / Dynamic Learning System
Paper # TL2011-29,NLC2011-26
Date of Issue

Conference Information
Committee NLC
Conference Date 2011/10/3(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 Refinement Method of Obtained Attributes for Undefined Concepts based on Temporal Alteration of Web
Sub Title (in English)
Keyword(1) Concept-base
Keyword(2) Auto Feedback
Keyword(3) Obtaining Attributes
Keyword(4) Dynamic Learning System
1st Author's Name Noriyuki OKUMURA
1st Author's Affiliation Department of Electronics and Computer Science()
Date 2011-10-10
Paper # TL2011-29,NLC2011-26
Volume (vol) vol.111
Number (no) 228
Page pp.pp.-
#Pages 4
Date of Issue