||Use of automatic keyphrase generation for creation of a construction thesaurus
||Kosovac B, Vanier D J
||The paper describes development of a thesaurus in the roofing domain. Thiswork is part of a larger effort to investigate the potential of thesauri as anaid in product modeling. Extractor, a software module that extractskeyphrases from documents, was used for collecting candidate thesaurusterms from Internet sources. The principal advantage of the Internet as asource of candidate terms is that it reflects colloquial language: -- thelanguage that is actually used by building practitioners and that it covers thewidest range of different 'user views' on the domain. The advantage of usingExtractor or similar software is that it allows processing huge text corporaavailable on the Internet and it eliminates irrelevant terms. Themethodology used was found to be highly useful, although it was notsufficient by itself for constructing a construction thesaurus, as considerablehuman intervention was required. Though limited time resources did notallow full exploitation of Extractor's capabilities, some possibilities forcustomization of the software and for partial automation of a thesaurusconstruction process are suggested.
|Year of publication:
Kosovac B, Vanier D J (1999).
Use of automatic keyphrase generation for creation of a construction thesaurus. Lacasse M A, Vanier D J (ed.); Information technology in construction, volume 4, ISBN 0-660-17743-9; Vancouver, May 30 - June 3, Canada (ISSN: 2706-6568),