Chinese word sense disambiguation using hownet
ICNC'05 Proceedings of the First international conference on Advances in Natural Computation - Volume Part I
| Hi-index | 0.00 |
Polysemy implies presence of more than one sense of aparticular word both in its context-bound and context-freesituation. The inherent aspect of polysemy is that a particularword will show multiple sense variations related by way ofsemantic extension and conceptual expansion. In last fiftyyears or so, polysemy is recognised as one of the centralissues in lexical semantics, word sense disambiguation (WSD),actual sense extraction (ASE), language learning, conceptualcategorisation of words as well as in computer processing oflanguage. Language users can identify a polysemous wordquite easily, but are not equipped enough to decipher all itspossible sense variations without appropriate reference toproper knowledge-base and other relevant informationembedded within contextual environments. Here we make anempirical effort to understand the basic nature of polysemy inBangla. We also intend to know how words denote sensevariations, which factors are instrumental in making thempolysemous, what impact do they create in languageunderstanding, and how sense variation can be bestunderstood using information from various sources ofknowledge base. Finally, we use a method to understand roleof various contexts, obtained from corpora, in maintaining aninterface between words and their sense variations. Quickreference to Local Context is handy at certain times, butreference to Focal, Topical and Global contexts as well as toextralinguistic knowledge base is necessary for understandingsense variation and for obtaining actual contextual sense.