Ant clustering algorithm with K-harmonic means clustering

Clustering is an unsupervised learning procedure and there is no a prior knowledge of data distribution. It organizes a set of objects/data into similar groups called clusters, and the objects within one cluster are highly similar and dissimilar with the objects in other clusters. The classic K-means algorithm (KM) is the most popular clustering algorithm for its easy implementation and fast working. But KM is very sensitive to initialization, the better centers we choose, the better results we get. Also, it is easily trapped in local optimal. The K-harmonic means algorithm (KHM) is less sensitive to the initialization than the KM algorithm. The Ant clustering algorithm (ACA) can avoid trapping in local optimal solution. In this paper, we will propose a new clustering algorithm using the Ant clustering algorithm with K-harmonic means clustering (ACAKHM). The experiment results on three well-known data sets like Iris and two other artificial data sets indicate the superiority of the ACAKHM algorithm. At last the performance of the ACAKHM algorithm is compared with the ACA and the KHM algorithm.