Extraction paramétrée de la terminologie du domaine
Mathieu Roche
Abstract
Automatic treatment of the specialyzed texts becomes more and more necessary due to their increasy number.
Term extraction, i.e., the extraction of groups of words significant for the field, is an information commonly required in the specialized domains.
In this paper, we propose a method for an automatic extraction of specific terms.
Our input is a corpus of specialized texts, upon which we carry out pretreatments : cleaning and labelling.
Next, we are using classical association measures to extract the terminology for the field.
Our main contribution is adding various parameters to improve the research for terms.
