A Novel Information Theoretic Approach for Finding Semantic Similarity in WordNet

Information Content (IC) based measures for finding semantic similarity is gaining preferences day by day as semantics of concepts can be highly characterized by information theory. This IC of concept can precisely quantify its generality and concreteness and generates dimensions for better understanding of concept semantics. The conventional way for calculating IC is based on the probability of appearance of concepts in corpora. Due to data sparseness and corpora dependency issues of those conventional approaches, a new corpora independent intrinsic IC calculation measure has evolved and gaining better performance over those conventional measures. In this paper we analyze several intrinsic IC models, emphasize related issues and present a novel information theoretic intrinsic model which can calculate IC of concepts based solely on underlying ontology. Our intense focus stays on several topological structures of the underlying ontology. Accuracy of intrinsic IC calculation measure relies on those factors deeply. Our approach is evaluated and compared with corpora and intrinsic IC based methods based on benchmark data set. Experimental results show that our intrinsic IC model achieves significant results than the existing techniques.