A Graph-based Clustering for Web Content Mining

摘要

In this paper we will introduce a new method of clustering where the data to be clustered will be represented by graphs instead of vectors or other models. Specifically, we will extend the classical kmeans clustering algorithm to work with graphs that represent web documents. We modeled web documents Content as graphs because they can allow us to retain information which is often discarded in simpler models. The experiments comparing the performance of clustering when using the traditional vector representation and our novel graph-based representations showed improvements in clustering quality and executing time over vector-based model.