An Effective Search Results Semantic Optimization Clustering Method for XML Fragments

With the emergence of more and more XML documents, the clustering of XML documents has become an active research area. However, it is more significance of XML element clustering than whole document due to the more focused and specific amount of information, especially to some application domain. Therefore, in this paper, we study the xml element clustering, in which the latent semantic indexing model is used to obtain the semantic relationship between terms firstly, and then a evaluation function for k-mediod is presented to automatic generate the optimal cluster number. In addition, information gain, for evaluating clustering quality is introduced. Experiment results show that the proposed semantic clustering optimization method outperforms the traditional method (No_optimization) in information gain and produces better clustering quality.