Using Multilingual Terms for Biomedical Term Extraction

The goal of automatic term extraction often is not so much the creation of a new list of domain specific terms, but rather the (semi-) automatic extension of a list of known terms. In this paper, we focus on the use of existing terms from glossaries, thesaurus, or ontologies to extract new terms from a domain specific text. Our new method is used to extract language-specific terms with the help of multilingual terminological resources. Our baseline system combines a linguistic pattern for extracting candidate noun phrases with a statistical method (χ) for ranking candidate phrases according to their association strength in a domain-specific corpus. Our scoring method also takes into account the termhood of candidate phrases computed on the basis of a list of known terms. We show that uninterpolated average precision of the resulting term list is improved when tested using human evalu-