Presentation 2008-06-30
Ontology-based Measuring of Semantic Similarity between Documents
Yumiko MIZOGUCHI, Toshiaki NAKAMOTO, Kazuma ASAKAWA, Shinichi NAGANO, Masumi INABA, Takahiro KAWAMURA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes techniques for measuring semantic similarity between documents. We use ontology to make a machine understand the meaning of a word. Our system measures similarity based on the distance between a pair of nodes. The words extracted from two documents for comparison are mapped to the nodes. In this paper, we focused on two processes. The first is a measuring similarity between a pair of nodes in ontology. The second is a method of aggregating the results of the similarity of each node. Human intuition is influenced not only by a distance between nodes but also by a structure of the ontology. In the ontology of the domain of the real world, the depth and width of a node's descendant are not uniform. Furthermore, a more important word influences human judgment more strongly. Our approach improved the correlation coefficient between the proposed approaches and human judgment by considering these human intuitions.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) semantic / similarity / ontology
Paper # AI2008-15
Date of Issue

Conference Information
Committee AI
Conference Date 2008/6/23(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 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) Ontology-based Measuring of Semantic Similarity between Documents
Sub Title (in English)
Keyword(1) semantic
Keyword(2) similarity
Keyword(3) ontology
1st Author's Name Yumiko MIZOGUCHI
1st Author's Affiliation Faculty of Engineering, Corporate Research & Development Center()
2nd Author's Name Toshiaki NAKAMOTO
2nd Author's Affiliation TOSHIBA INFORMATION SYSTEMS (JAPAN) CORPORATION
3rd Author's Name Kazuma ASAKAWA
3rd Author's Affiliation TOSHIBA INFORMATION SYSTEMS (JAPAN) CORPORATION
4th Author's Name Shinichi NAGANO
4th Author's Affiliation Faculty of Engineering, Corporate Research & Development Center
5th Author's Name Masumi INABA
5th Author's Affiliation Faculty of Engineering, Corporate Research & Development Center
6th Author's Name Takahiro KAWAMURA
6th Author's Affiliation Faculty of Engineering, Corporate Research & Development Center
Date 2008-06-30
Paper # AI2008-15
Volume (vol) vol.108
Number (no) 119
Page pp.pp.-
#Pages 6
Date of Issue