A methodology for selecting the most suitable cluster validation internal indices

Abstract

Validation of clustering results is an important issue in the context of machine learning research and it is essential for the success of clustering applications. Choosing the appropriate validation index for evaluating the results of a particular clustering algorithm remains a challenge. The quality of partitions generated by different clustering algorithms can be evaluated using different indices based on external or internal criteria. In this paper, we have proposed a methodology for selecting the most suitable cluster validation internal index, relating external and internal criteria through a regression model applied on the results of partitioning clustering algorithm.