reviewed two clustering algorithms (hard c-means [16] and single linkage) and three indexes of crisp cluster validity (Huberts statistics, the Davies-Bouldin index, and Dunns index) [17]. Their work illustrates two deficiencies of Dunns index [18] that make it overly sensitive to noisy clusters. They proposed several generalizations of those deficiencies which are not as brittle to outliers in the clusters. Definitions regarding cluster in a graph and measures of cluster quality were reviewed in [10]. This work also presented global algorithms for clustering the entire vertex set of an input graph and discussed the task of identifying a cluster for a specific seed vertex by local computation.