Classifying Domain-Specific Terms Using a Dictionary

Classifying Domain-Specific Terms Using a Dictionary Su Nam Kim Dept. of CSSE University of Melbourne Melbourne, Australia snkim@csse.unimelb.edu.au Lawrence Cavedon School of CS IT RMIT University Melbourne, Australia lawrence.cavedon@rmit.edu.au Abstract Automatically building domain-specific on- tologies is a highly challenging task as it re- quires 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 pre- vious approaches in related research (such as word sense disambiguation (WSD) and named entity recognition (NER)) that use semantic similarity among domain concepts, we ex- plore 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 domain- specific terms from FOLDOC, a freely avail- able on-line dictionary for the the Comput- ing domain, and defined 9 domain concepts for this domain. Our results show that be- yond contextual features, using domain con- cepts and topics derived from domain-specific terms helps to improve assigning domain con- cepts to the terms. 1 Introduction Domain-specific terms are terms that have signif- icant meaning(s) in a specific domain. For exam- ple, terms such as Gulf and Kuwait are associated with the domain of oil due to their frequent appear- ances in contexts related to oil although they indi- cate geographical areas. In some resources, domain- specific terms are further categorized in terms of their domain concepts (i.e., semantic labels/classes). For example, Firefox belongs to the domain con- cept Software, while Prolog is associated with the domain concept Programming. In this paper, we use the term domain concept for consistency. Note that in previous work, the meaning of the domain- specificity is associated with either word senses (e.g. (Magnini et al., 2002; Rigutini et al., 2005)) or the statistical use of terms in context (e.g. (Drouin, 2004; Milne et al., 2006; Kida et al., 2007; Park et al., 2008; Kim et al., 2009; Vivaldi and Rodr- guez, 2010)). In WordNet, the domain concept is assigned based on the word senses. Similarly, WordNet Domain has terms with domain con- cepts per sense. However, most work previously conducted work used domain-specificity is based on statistical use. In this paper, we follow the latter def- inition, i.e., domain-specificity associated with the statistical use of the term. Domain-specificity of terms has been leveraged in various natural language processing (NLP) and related tasks, such as word sense disambiguation (WSD) (Magnini et al., 2002), named entity recog- nition (NER) (Kazama and Torisawa, 2007), and query expansion (Rigutini et al., 2005). Resources containing domain information fall into two groups: the list of domain-specific terms without domain concepts (e.g. Agrivoc, EUROVOC, ASFA The- saurus); and with domain concepts (e.g. WordNet, WordNet Domain, Unified Medical Language Sys- tem (UMLS)). Although there have been efforts de- veloping such knowledge resources, the task has been generally carried out by hand, requiring high cost and time. Further, even hand-crafted …

Content truncated to 500 words