University of Western Sydney School of Computing and Information Technology Cluster Validity Using Support Vector Machines

Gaining confidence that a clustering algorithm has produced meaningful results and not an accident of its usually heuristic optimization is central to data analysis. This is the issue of validity and we propose here a method by which Support Vector Machines are used to evaluate the separation in the clustering results. However, we not only obtain a method to compare clustering results from different algorithms or different runs of the same algorithm, but we can also filter noise and outliers. Thus, for a fixed data set we can identify what is the most robust and potentially meaningful clustering result. A set of experiments illustrates the steps of our approach.