A General K-Mean Clustering Algorithm Based On Constrained Dynamic Time Warping Distance Measure

Abstract

Clustering is a division of data into groups of similar objects. Each group, called cluster, consists of objects that are similar between themselves and dissimilar to objects of other groups. In high dimensional data space, clusters are likely to exist in different subspaces. General K-Mean (GKM) is a classic clustering algorithm, but it cannot be used to find subspace clusters. In this work, Dynamic Time Warping (DTW) is a much more dynamic distance measure for time series, allowing comparable shapes to competition even this work is out of phase in the time association. It permits a non-linear illustration of single suggestion to a different by reducing the space among the two. A decade back, DTW was establishing into Data Mining neighborhood as effectiveness for different responsibilities for moments sequence evils including categorization, group, and variance discovery. Experimental results make obvious that the DTW advances create better performance than GKM clustering algorithms.