Analysis of Inter cluster movement based on geometric probability and regression

Abstract

This paper proposes an approach to study the behavior of clustering system, when unclustered data comes to the existing clustered setup. With the introduction of a new data point to the system, there is a possible occurrence of migration of points between clusters, resulting in re-clustering of the setup. An attempt was made to assess the inter cluster movement scenario, with geometric probabilistic approach and regression model, studying the possibility of migration initiated by the new data point which may be located at varying distance from the cluster center and simultaneously increasing the size of the clusters and dimension of the clustering data. A comparative study on the results obtained by both these models was framed and the results reveal lower relative error, which increases with increasing size and the dimension of the clustering system.