Building a large thesaurus for information retrieval
ANLC '88 Proceedings of the second conference on Applied natural language processing
| Hi-index | 0.00 |
Machine readable dictionaries may be invaluable components of future in-formation retrieval systems. The Virginia Tech department of Computer Science is fortunate to have available for research purposes machine readable versions of the Collins English Dictionary (CED) [HANK 79] and of the Oxford Advanced Learners Dictionary of Current English (OALDCE) [HORN 74] among others. McIllroy [MCIL 84] and Mitton [MIlT 85] have put forth separate efforts in cleaning up the OALDCE. The CED is only available in its original typesetting form, and it required a great deal of effort and patience to transform this into a relational lexicon. A number of recent efforts have been made to generate relational lexicons from dictionaries (e.g., [AHLS 81], [AHLS 83], [AHLS 85], [WHIT 83], [EVEN 78] ). The lexicon that will be discussed in this paper consists of a large set of relations in the form of PROLOG facts. This knowledge base will become a very important entity in the analysis and retrieval sub-systems of the CODER project [FOXE 86].