Output details

Learning semantic relatedness from term discrimination information

<17> This paper explores and develops a novel method for estimating the discrimination information conveyed by terms using the information radius measure. The paper then formally defines and mathematically interprets discrimination measures and, in particular, argues that the interpretation is essential for practical applications. The paper discusses important features of, and differences between, the information radius measure and other information measures from an application perspective. The rigour of the paper lies in a substantial concept and property integration and reflection through practical examples. ESwA is a leading journal in applied Computing Science with an impact factor of 2.91 in 2009.