Building Inter Cluster Movement Estimation (ICME) model - A step by step approach

Citations of this article
Mendeley users who have this article in their library.


One of the most well-known techniques in data mining is clustering. This paper presents a scenario of introducing new unclustered information to the already clustered system. Consequently, there is an occurrence of movement of data points between clusters, to accommodate the new entrée. This paper attempts to develop an Inter Cluster Movement Estimation (ICME) model to predict this behaviour of the data points in the clustering system. Better prediction will result in the reduction in the number of times, re-clustering is done on the data sets. Experiments were conducted on datasets with multiple instances and attributes. On analysis, the study revealed that ICME model was in concurrence with observed values with a lower error rate. Real data sets from UCI Data Repository were used for comparative analysis of ICME with similar partitional clustering algorithms including adaptive K Mean and Fuzzy C Means. Reports prove that ICME was found to converge faster consuming lesser number of iterations than adaptive K means and Fuzzy C Means.




Rajee, A. M., & Sagayaraj Francis, F. (2015). Building Inter Cluster Movement Estimation (ICME) model - A step by step approach. In Procedia Computer Science (Vol. 46, pp. 210–215). Elsevier B.V.

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free