Acquiring contextual information for term disambiguation
Diana Maynard, S Ananiadou · Research Explorer (The University of Manchester) · 1998
Term extraction is important for many information systems applications. Although terms should be monoreferential, they actually exhibit a high degree of ambiguity. This paper describes a method for acquiring contextual information for term disambiguation. Unlike traditional statistical techniques, semantic information about the context is acquired from a domain-specific thesaurus, in order to define relationships between context and term. Using this information, we evaluate the significance of context words with respect to their co-occurring terms. 1 Introduction The increasing volumes of electronic textual data in recent years have given rise to a need for term extraction applications, for example in dictionary construction, automatic indexing, machine translation, information retrieval and hypertext linking. The fact that terms exhibit a high degree of ambiguity can be problematic for these applications. Traditional methods of word sense disambiguation rely on statistical methods, ...