Classifying Domain-Specific Terms Using a Dictionary

Automatically building domain-specific ontologies is a highly challenging task as it requires extracting domain-specific terms from a corpus and assigning them relevant domain concept labels. In this paper, we focus on the second task: i.e., assigning domain concepts to domain-specific terms. Motivated by previous approaches in related research (such as word sense disambiguation (WSD) and named entity recognition (NER)) that use semantic similarity among domain concepts, we explore three types of features — contextual, domain concepts, topics — to measure the semantic similarity of terms; we then assign the domain concepts from the best matching terms. As evaluation, we collected domainspecific terms from FOLDOC, a freely available on-line dictionary for the the Computing domain, and defined 9 domain concepts for this domain. Our results show that beyond contextual features, using domain concepts and topics derived from domain-specific terms helps to improve assigning domain concepts to the terms.