Anzeige
Erschienen in:
2016 | OriginalPaper | Buchkapitel
Measurement of Semantic Similarity: A Concept Hierarchy Based Approach
verfasst von : Shrutilipi Bhattacharjee, Soumya K. Ghosh
Erschienen in: Proceedings of 3rd International Conference on Advanced Computing, Networking and Informatics
Verlag: Springer India
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
Resolving semantic heterogeneity is one of the major issues in many fields, namely, natural language processing, search engine development, document clustering, geospatial information retrieval and knowledge discovery, etc. Semantic heterogeneity is often considered as an obstacle for realizing full interoperability among diverse datasets. Appropriate measurement metric is essential to properly understand the extent of similarity between concepts. The proposed approach is based on the notion of concept hierarchy which is built using a lexical database. The WordNet, a semantic lexical database, is used here to build the semantic hierarchy. A measurement metric is also proposed to quantify the extent of similarity between a pair of concepts. The work is compared with existing methodologies on Miller-Charles benchmark dataset using three correlation coefficients (Pearson’s, Spearman’s and Kendall Tau rank correlation coefficients). The proposed approach is found to yield better results than most of the existing techniques.