IDENTIFYING DOMAIN-SPECIFIC SENSES AND ITS APPLICATION TO TEXT CLASSIFICATION
Fumiyo Fukumoto, Yoshimi Suzuki
2010
Abstract
This paper focuses on domain-specific senses and presents a method for identifying predominant sense depending on each domain. The method consists of two steps: selecting senses by text classification and scoring senses by link analysis. Sense selection is to identify each sense of a word to the corresponding domain. We used a text classification technique. Senses were scored by computing the rank scores using the Markov Random Walk (MRW) model. The method was tested on WordNet 3.0 and the Reuters corpus. For evaluation of the method, we compared the results with the Subject Field Codes resources, which annotate WordNet 2.0 synsets with domain labels. Moreover, we applied the results to text classification. The results demonstrated the effectiveness of the method.
DownloadPaper Citation
in Harvard Style
Fukumoto F. and Suzuki Y. (2010). IDENTIFYING DOMAIN-SPECIFIC SENSES AND ITS APPLICATION TO TEXT CLASSIFICATION . In Proceedings of the International Conference on Knowledge Engineering and Ontology Development - Volume 1: KEOD, (IC3K 2010) ISBN 978-989-8425-29-4, pages 263-268. DOI: 10.5220/0003094102630268
in Bibtex Style
@conference{keod10,
author={Fumiyo Fukumoto and Yoshimi Suzuki},
title={IDENTIFYING DOMAIN-SPECIFIC SENSES AND ITS APPLICATION TO TEXT CLASSIFICATION},
booktitle={Proceedings of the International Conference on Knowledge Engineering and Ontology Development - Volume 1: KEOD, (IC3K 2010)},
year={2010},
pages={263-268},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003094102630268},
isbn={978-989-8425-29-4},
}
in EndNote Style
TY - CONF
JO - Proceedings of the International Conference on Knowledge Engineering and Ontology Development - Volume 1: KEOD, (IC3K 2010)
TI - IDENTIFYING DOMAIN-SPECIFIC SENSES AND ITS APPLICATION TO TEXT CLASSIFICATION
SN - 978-989-8425-29-4
AU - Fukumoto F.
AU - Suzuki Y.
PY - 2010
SP - 263
EP - 268
DO - 10.5220/0003094102630268