Improving text categorization by resolving semantic ambiguity

Authors:
Hiroshi Uejima;Takao Miura;Isamu Shioya
Affiliations:
Department of Electrical and Electronic Engineering, Hosei University, Koganei, 184-8584 Japan;Department of Electrical and Electronic Engineering, Hosei University, Koganei, 184-8584 Japan;Department of Management and Informatics, Sanno University, Isehara, 259-1197 Japan
Venue:
Systems and Computers in Japan
Year:
2005

Citing 0
Cited 1

The centrality of user modeling to high recall with high precision search

SMC'09 Proceedings of the 2009 IEEE international conference on Systems, Man and Cybernetics

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this investigation, we propose a new method for text categorization (TC) based on a Bayesian approach with resolution of ambiguity. TC assigns weights to words whose meanings are ambiguous in the sense of synonymy and polysemy. We give weights to articles by examining dictionaries of thesaurus type and use dimensionality reduction to improve the quality of TC. We also utilize WordNet as a lexical reference tool and present some experiments to illustrate the effectiveness of our approach. © 2005 Wiley Periodicals, Inc. Syst Comp Jpn, 36(4): 1–8, 2005; Published online in Wiley InterScience (). DOI 10.1002/scj.20191