Clustering XML Documents Based on the Weight of Frequent Structures

The previous clustering methods of XML document group XML documents with similar structures, measuring structural similarity and distance between XML documents. In this paper, however, we propose a novel clustering method for XML documents using the weight of frequent structures in XML documents, considering that an XML document as a transaction and the extracted structures from XML documents as items of a transaction. Our experiment results show the high speed and cluster cohesion of our clustering method.