Abstract
The rapid development of science and technology in different domains has created many new concepts and the domain lexicon must be updated timely to include the new terms as domain knowledge. However, automatic update of domain knowledge requires a core lexicon for bootstrapping purpose. The core lexicon should contain the fundamental terms used in a domain and from the core lexicon other concepts and terms can be built upon. In this paper we present an algorithm for extracting the core lexicon from some domain specific lexicons. Experiment on a large domain specific lexicon with 139,429 entries shows that only 3,413 terms form the core lexicon with a high precision of 97% and a good coverage.
Original language | English |
---|---|
Title of host publication | Proceedings - ALPIT 2007 6th International Conference on Advanced Language Processing and Web Information Technology |
Pages | 183-188 |
Number of pages | 6 |
DOIs | |
Publication status | Published - 1 Dec 2007 |
Event | 6th International Conference on Advanced Language Processing and Web Information Technology, ALPIT 2007 - Luoyang, Henan, China Duration: 22 Aug 2007 → 24 Aug 2007 |
Conference
Conference | 6th International Conference on Advanced Language Processing and Web Information Technology, ALPIT 2007 |
---|---|
Country/Territory | China |
City | Luoyang, Henan |
Period | 22/08/07 → 24/08/07 |
ASJC Scopus subject areas
- General Computer Science
- Information Systems