Word Sense Disambiguation of semantic document

Abstract

A Max-Probability Density based Clustering (MPDC) algorithm is proposed in this paper to resolve the problem of Word Sense Disambiguation in semantic document. MPDC take the context information of a keyword based on WordNet into account and select the max probability sense by measuring the density of the concept. We also do experiment on semantic documents retrieving from Swoogle and Watson, two famous semantic web searching engines. The result shows MPDC get a good efficiency.