WI-IAT '09 Proceedings of the 2009 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology - Volume 03
Resources for Turkish morphological processing
Language Resources and Evaluation
Hi-index | 0.00 |
This paper proposes a method of domain classification oftechnical terms using the Web. In the proposed method, it isassumed that, for a certain technical domain, a list of knowntechnical terms of the domain is given. Technical documents of thedomain are collected through the Web search engine, which are thenused for generating a vector space model for the domain. The domainspecificity of a target term is estimated according to thedistribution of the domain of the sample pages of the target term.Experimental evaluation results show that the proposed method ofdomain classification of a technical term achieved mostly90% precision/recall. We then apply thistechnique of estimating domain specificity of a term to the task ofdiscovering novel technical terms that are not included in anyexisting lexicons of technical terms of the domain. Out of 1000randomly selected candidates of technical terms per domain, wediscovered about 100 to 200 novel technical terms. © 2007Wiley Periodicals, Inc. Syst Comp Jpn, 38(14): 1119, 2007;Published online in Wiley InterScience (). DOI10.1002/scj.20852