CLUSTER VALIDATION USING GRAPH-THEORETIC CONCEPTS

In this article we have generalized Dunn's index and the Davies-Bouldi n index for cluster validation using graph structures, such as GG, RNG and MST. Unlike Dunn's index and the Davies-BouIdin index, the propos ed indices are not sensitive to noisy points and are applicable to hyp erspherical and structural clusters as well. The relationships between Various indices have also been established. The effectiveness of the generalized indices and superiority over some existing cluster validit y indices are established using eight data sets. (C) 1997 Pattern Reco gnition Society.