An Automatic Extension Method of Japanese-WordNet by Using Wikipedia Category Hierarchy

Akio KOBAYASHI  Shigeru MASUYAMA 

Publication
D - Abstracts of IEICE TRANSACTIONS on Information and Systems (Japanese Edition)  Vol.J95-D  No.6  pp.1356-1368
Publication Date: 2012/06/01
Online ISSN: 1881-0225
Print ISSN: 1880-4535
Type of Manuscript: PAPER
Category: 
Keyword: 
ontologies,  information extraction,  text mining,  knowledge representation, 

Full Text(in Japanese): PDF(781.2KB)

Summary: 
Japanese-WordNet is a license free large thesaurus, which is a Japanese version of Princeton WordNet. However, Japanese-WordNet does not include proper nouns except popular nouns such as a nation, which causes some problems in analyzing proper nouns or new words. In contrast, Wikipedia, the online encyclopedia has a large amount of proper nouns and new words as entries, which are increasing and updated every day. However, Wikipedia does not have a highly accurate word classification system like Wordnet. Thus, we propose a method for connecting Wikipedia category hierarchies to a semantic class of Japanese-Wordnet for extending it.