Rough set theory is a powerful mathematical tool that has been applied widely to extract knowledge from many databases. It deals with inexact and incomplete data. Cluster analysis means finding similarities between data according to the characteristics found in the data and grouping similar data objects into clusters that are meaningful, useful, or both. Data clustering techniques are valuable tools for researchers working with large databases of multivariate data. Cluster analysis is used in various applications viz., Pattern Recognition, Data Analysis, Image Processing and so on. This paper analyses Roughness and Maximum Dependency Attribute clustering algorithms that minimizes the need for subjective human intervention and compare the purity analysis between these two methods. Purity analysis percentage is calculated from the result of final clusters. Six datasets are used in this research work for comparing the roughness and maximum dependency attribute algorithm to describe the cluster solution by using the purity analysis (PA).

Rough Clustering, Equivalence Classes, Roughness, Maximum Dependency Attribute, Purity Analysis

[1] Arun K Pujari, “Data Mining Techniques”, Universities Press (India) Private Limited, 2010.
[2] C. L. Blake and C.J. Merz, “UCI Repository of Machine Learning Databases”, Irvin, University of California, http://www.ics.uci.eduction."/~mlearn/, 1998.
[3] Chandranath Adak, “Rough Clustering Based Unsupervised Image Change Detection”.
[4] J.Han and M.Kamber, “Datamining: Concepts and Techniques”, Morgan Kaufmann Publishers, 1992.
[5] Jun-Hao Zhang, Ming-Hu Ha, Jing Wu, “Implementation of Rough K-Means Clustering Algorithm in MATLAB”, 9th International Conference on Machine Learning and Cybernetics, pp.2084-2087, July 2010.
[6] Manish Joshi, Yiyu Yao, Pawan Lingras, Virendrakumar C.Bhavsar, “Rough, Fuzzy, Interval Clustering For Web Usage Mining”, 10th International Conference on Intelligent Systems Design and Applications, pp.397- 402, 2010.
[7] Pawan Lingras, Rui Yan, Chad West, “Comparison of Conventional And Rough K-Means Clustering”, LNAI 2639, Springer – Verlag Berlin Heidelberg,pp.130-137,2003.
[8] Z. Pawlak, “Rough sets”, International Journal of Computer and Information Sciences, vol. 11, pp. 341–356, 1982.
[9] Pethalakshmi.A, A.Banumathi, “Refinement of K-Means And Fuzzy C-Means”, International Journal of Computer Applications, Volume 39, Paper Number: 17, Feb 2012.
[10] Pethalakshmi.A, A.Banumathi, “A Novel Approach in Clustering via Rough set”, IJSR, vol. 2, Issue 7, July 2013.
[11] Tien-Chin Wang, Lisa Y.Chen, Hsien-Da Lee, “Fuzzy Entropy-Based
Rough Set Approach for Extracting Decision Rules”, pp.5636-5639, IEEE 2007.