3M algorithm: finding an optimal fuzzy cluster scheme for proximity data
Ying Xie, Vijay V. Raghavan, Xiaoquan Zhao · 2003
In order to find an optimal fuzzy cluster scheme for proximity data, where just pairwise distances among objects are given, two conditions are necessary: A good cluster validity function, which can be applied to proximity data for evaluation of the goodness of cluster schemes for varying number of clusters; a good cluster algorithm that can deal with proximity data and produce an optimal solution for a fixed number of clusters. To satisfy the first condition, a new validity function is proposed, which works well even when the number of clusters is very large. For the second condition, we give a new algorithm called multi-step maxmin and merging algorithm (3M algorithm). Experiments show that, when used in conjunction with the new cluster validity function, the 3M algorithm produces satisfactory results.