The fuzzy k-means clustering algorithm is another overlapping clustering algorithm. It is an extension of k-means algorithm without the restriction of exclusive clusters. One data point can be a part of more than one cluster. In the overlapping clusters, any point can belong to more than one cluster with a certain affinity value toward each cluster. This affinity is proportional to the distance from the point to the centroid of the cluster. Fuzzy k-means converges faster than k-means, and should be preferred if the criteria of exclusivity is not mandatory.
The fuzzy k-means algorithm has a parameter, m, called the fuzziness factor. Like k-means, fuzzy k-means loops over the dataset. However, instead of assigning vectors to the nearest centroids, it calculates the degree of association of the point to each of the clusters.
The fuzzy k-means algorithm starts behaving more like the k-means algorithm as m gets closer to 1. If m increases, the fuzziness of...
