Semantic Similarity Measures

Semantic similarity measures can help computers to distinguish one object form another, group them based on the similarity, classify a new object into the group, or even predict the behavior of the new object. During recent years several attempts have been made in using different semantic similarity measures for information retrieval purposes. And the truth is that there are a substantial number of distance/similarity measures encountered in many different fields such as anthropology, biology, chemistry, computer science, ecology, information theory, geology, mathematics, physics, psychology, statistics, etc.

Our research interest is focused on semantic similarity measures rather than semantic relatedness or semantic distance, which are also often used in the technical literature. For example, in their extensive survey of relatedness measures, Budanitsky and Hirst [2006] argued that the notion of relatedness is more general than that of similarity, as the former subsumes many different kinds of specific relations, including meronymy, antonymy, functional association, and other approaches.