New K-means Clustering Method Using Minkowski’s Distance as its Metric

Cluster analysis is an unsupervised learning method that classifies data points, usually multidimensional into groups (called clusters) such that members of one cluster are more similar (in some sense) to each other than those in other clusters. In this paper, we propose a new k-means clustering method that uses Minkowski’s distance as its metric in a normed vector space which is the generalization of both the Euclidean distance and the Manhattan distance. The k-means clustering methods discussed in this paper are Forgy’s method, Lloyd’s method, MacQueen’s method, Hartigan and Wong’s method, Likas’ method and Faber’s method which uses the usual Euclidean distance. It was observed that the new k-means clustering method performed favourably in comparison with the existing methods in terms of minimization of the total intra-cluster variance using simulated data and real-life data sets.