A Hybrid Approach for Measuring Semantic Similarity between Documents and its Application in Mining the Knowledge Repositories

Abstract

—This paper explains about similarity measure and the relationship between the knowledge repositories. This paper also describes the significance of document similarity measures, algorithms and to which type of text it can be applied Document similarity measures are of full text similarity, paragraph similarity, sentence similarity, semantic similarity, structural similarity and statistical measures. Two different frameworks had been proposed in this paper, one for measuring document to document similarity and the other model which measures similarity between documents to multiple documents. These two proposed models can use any one of the similarity measures in implementation aspect, which is been put forth for further research.