Efficiency of k-Means and K-Medoids Algorithms for Clustering Arbitrary Data Points

There are number of techniques proposed byseveral researchers to analyze the performance ofclustering algorithms in data mining. All thesetechniques are not suggesting good results for thechosen data sets and for the algorithms in particular.Some of the clustering algorithms are suit for somekind of input data. This research work usesarbitrarily distributed input data points to evaluatethe clustering quality and performance of two of thepartition based clustering algorithms namely k-Means and k-Medoids. To evaluate the clusteringquality, the distance between two data points aretaken for analysis. The computational time iscalculated for each algorithm in order to measure theperformance of the algorithms. The experimentalresults show that the k-Means algorithm yields thebest results compared with k-Medoids algorithm.