| Abstract |
ln this paper, we describe a method to build terminological selection filters using an existing terminological resource, the EDF thesaurus. We first proceed to a linguistic enrichment of the available terminology, using NLP software and an existing dictionary. This process leads us to manipulate terms as decorated syntactic trees. Then we try to model what a term is in the EDF thesaurus only by considering the relationship between nouns and adjectives, using a relational analysis clusterization software from IBM. The result is the induction of rich patterns, based on morphological, syntactic, semantic and domain information. These patterns help to filter and select term candidates from a text, and so, make easier the updating of our thesaurus. |
| BibTex |
@InProceedings{ELX96_2-044,
author = {Elie Naulleau, Marie-Gaëlle Monteil, Benoît Habert},
title = {Recycling a Thesaurus to Characterize and Process Terms},
pages = {807-816},
booktitle = {Proceedings of the 7th EURALEX International Congress},
year = {1996},
month = {aug},
date = {13-18},
address = {Göteborg, Sweden},
editor = {Martin Gellerstam, Jerker Järborg, Sven-Göran Malmgren, Kerstin Norén, Lena Rogström, Catalina Röjder Papmehl},
publisher = {Novum Grafiska AB},
isbn = {91-87850-14-1},
} |
