A new robust validity index for fuzzy clustering algorithm
Horng-Lin Shieh, Po-Lun Chang · 2010
This paper proposes a robust validity index for Fuzzy c-Means (FCM) algorithm. The Fuzzy c-Means algorithm has become of most widely used method in fuzzy clustering. After clustering, it is often necessary to evaluate its results. Such assessment techniques are called cluster validity. The disadvantage of FCM is that the number of clusters must be predetermined. Even if the number of clusters is given, the clustering results of these algorithms are influenced by the choice of initial cluster centers. In this paper, a new cluster validity index is proposed to evaluate the fitness of clusters obtained by FCM. The example shows the result of proposed index have good performances than other cluster validities.