Presentation 1999/1/12
Measuring Semantic Similarity Model between Concepts for Intelligent Mechanism
Takeshi Irie, Hirokazu Watabe, Tsukasa Kawaoka, Kazumitsu Matsuzawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The basic factor of information processing systems which have friendly user interfaces is a intelligent mechanism, and the main elements of that are the concept-base and the association mechanism using similarity between concepts in the concept-base. A structure of the concept-base must be simple since the concept-base can be maintained easily. This paper proposes measuring semantic similarity model between concepts based on such a conditions of the concept-base. The feature of proposed model is not that the degree of similarity evaluates by difference of first stage attributes vector but the degree of similarity evaluates by set of second stage attributes as semantic points. By experimental results using the concept-base which consists of about 40, 000 concepts, it is shown that the proposed model can output the closer degree of similarity to that decided by human judges.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) concept-base / chain concept / measuring semantic similarity / view point / n-stage attribute
Paper # AI98-75
Date of Issue

Conference Information
Committee AI
Conference Date 1999/1/12(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 Artificial Intelligence and Knowledge-Based Processing (AI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Measuring Semantic Similarity Model between Concepts for Intelligent Mechanism
Sub Title (in English)
Keyword(1) concept-base
Keyword(2) chain concept
Keyword(3) measuring semantic similarity
Keyword(4) view point
Keyword(5) n-stage attribute
1st Author's Name Takeshi Irie
1st Author's Affiliation Department of Knowledge Engineering and Computer Science, Faculty of Engineering, Doshisha University()
2nd Author's Name Hirokazu Watabe
2nd Author's Affiliation Department of Knowledge Engineering and Computer Science, Faculty of Engineering, Doshisha University
3rd Author's Name Tsukasa Kawaoka
3rd Author's Affiliation Department of Knowledge Engineering and Computer Science, Faculty of Engineering, Doshisha University
4th Author's Name Kazumitsu Matsuzawa
4th Author's Affiliation NTT Communication Science Laboratories
Date 1999/1/12
Paper # AI98-75
Volume (vol) vol.98
Number (no) 499
Page pp.pp.-
#Pages 8
Date of Issue