Using Exponential Kernel for Word Sense Disambiguation

Tinghua Wang,Junyang Rao,Dongyan Zhao
DOI: https://doi.org/10.1007/978-3-642-40728-4_68
2013-01-01
Abstract:The success of machine learning approaches to word sense disambiguation (WSD) is largely dependent on the representation of the context in which an ambiguous word occurs. Typically, the contexts are represented as the vector space using \"Bag of Words (BoW)\" technique. Despite its ease of use, BoW representation suffers from well-known limitations, mostly due to its inability to exploit semantic similarity between terms. In this paper, we apply the exponential kernel, which models semantic similarity by means of a diffusion process on a graph defined by lexicon and co-occurrence information, to smooth the BoW representation for WSD. Exponential kernel virtually exploits higher order co-occurrences to infer semantic similarities in an elegant way. The superiority of the proposed method is demonstrated experimentally with several SensEval disambiguation tasks.
What problem does this paper attempt to address?