Clustering of dissimilar perception phase constructed for similarity measures using k-means algorithm

Abstract

The process of grouping a set of objects into classes of Similar objects is called clustering. Clustering definition differs from one model to another, but in most of the clustering models the similarity concept is based on distances, like Euclidean distance. In this paper an attempt is made to explore a general type of Similarity like multi viewpoint based on similarity measures and Related two clustering methods. The difference between similarity and dissimilarity is that the earlier clustering technique in which the distinct opinion which is the origin, by keeping this origin the cluster objects were assumed to be present in the same cluster by measuring the two different objects. Clustering method is used for verdict similarity sets in a bigdata, called clusters. By using this technique an attempt to cluster individuals in a population together by similarity is done. Clustering sometimes called an unsupervised learning, since the data have not been prescribed by any index to the data and no class values representing a priori grouping of the instances of data given.. In this paper clustering technique is applied and tried to apply the same by using the centroid based clustering algorithm and K means techniques.