Computational time factor analysis of K-means algorithm on actual and transformed data clustering

Abstract

Clustering is the process of partitioning a set of objects into a distinct number of groups or clusters, such that objects from the same group are more similar than objects from different groups. Clusters are the simple and compact representation of a data set and are useful in applications, where we have no prior knowledge about the data set. There are many approaches to data clustering that vary in their complexity and effectiveness due to its wide number of applications. K-means is a standard and landmark algorithm for clustering data. This multi-pass algorithm has higher time complexity. But in real time we want the algorithm which is time efficient. Hence, here we are giving a new approach using wiener transformation. Here the data is wiener transformed for k-means clustering. The computational results shows that the proposed approach is highly time efficient and also it finds very fine clusters.