Ensemble based distributed k-harmonic means clustering

Abstract—Due to the explosion in the number of autonomous data sources, there is a growing need for effective approaches for distributed knowledge discovery and data mining. The distributed clustering algorithm is used to cluster the distributed datasets without necessarily downloading all the data to a single site. K-Means is used as a popular clustering method due to its simplicity and high speed in clustering large datasets. The dependency of the K-Means performance on the initialization of centroids is a major problem. Similarly, distributed clustering algorithm based on K-Means is also sensitive to centroid initialization. It is demonstrated that K-Harmonic Means is essentially insensitive to centroid initialization. In this paper, a novel ensemble based distributed clustering algorithm using K-Harmonic Means is proposed. The simulated experiments described in this paper confirm robust performance of the proposed algorithm