Evaluate the Performance of the Clustering Algorithms by Using Data Discrepancy Factor


S Govinda Rao,N V Ganapathi Raju,A Sai Hanuman,P Varaprasada Rao,




K-Means,Modified K-Means,Hierarchical Clustering,DDF,Modern DDF,


DDF is the most valuable measure among various cluster performance techniques to evaluate the perfectness of any cluster mechanism. Normally, best clusters are evaluated by computing the number of data points within a cluster. When this count is equivalent to the number of required data points then this cluster is considered to be perfect. The excellence of the cluster methodology is essential not only to find the data count inside a cluster but also to examine it by totaling the data points these are (i) present within a cluster where it should not be and vice versa and (ii) not clustered i.e. outliers (OL). The main functionality of DDF is that all cluster points can be grouped in similar clusters without outliers, the present paper highlights on how compared to DDF more efficient Clusters can be formed through the Modern DDF. Further, we evaluate the performance of some clustering algorithms, K-Means. Recently we developed the Modified K-Means Algorithm and Hierarchical Algorithm by using the Data Discrepancy Factor (DDF).


I. B.Giovanni, “AClAP, Autonomous hierarchical agglomerative Cluster
Analysis based protocol to partition conformational datasets.” Bioinformatics
Vol: 22, Issue: 14, pp: e58-e65, 2006.
II. M.Ujjwal, S.Bandyopadhyay. “Performance evaluation of some clustering
algorithms and validity indices.” IEEE Transactions on Pattern Analysis and
Machine Intelligence Vol:24, Issue: 12, pp: 1650-1654, 2002.
III. N.Shi, L.Xumin, G.Yong. “Research on k-means clustering algorithm: An
improved k-means clustering algorithm.”Intelligent Information Technology
and Security Informatics (IITSI), Third International Symposium on. IEEE,
IV. O.J.Oyelade, , O.Oladipupo, I.C.Obagbuwa. “Application of k Means
Clustering AlgorithmFor prediction of Students Academic Performance.”
arXiv preprint arXiv:1002.2425, 2010.
V. R.P.Vaishali, R.G.Mehta. “Modified k-means clustering algorithm.”
Computational Intelligence and Information Technology. Springer, Berlin,
Heidelberg, pp: 307-312, 2011.
VI. S.E.Brian, “Hierarchical clustering.” Cluster Analysis, 5th Edition, pp: 71-
110, 2011.

VII. S.G.Rao, A.Govardhan. “Assessing h-and g-Indices of Scientific Papers using
k-MeansClustering.” International Journal of Computer Applications Vol:
100, Issue: 11, 2014.
VIII. S.G.Rao, A.Govardhan. “Investigation of Validity Metrics for Modified KMeans
Clustering Algorithm.” i-Manager’s Journal on Computer Science
Vol: 3, Issue: 2, pp: 33, 2015.
IX. S.G.Rao, A.Govardhan. “Performance Validation of the Modified K-Means
Clustering Algorithm Clusters Data.” International Journal of Scientific &
Engineering Research Vol: 6, Issue: 10, pp: 726-730, 2015.
X. X.Juanying, “An Efficient Global K-means Clustering Algorithm.” JCP
Vol:6, Issue: 2, pp:271-279, 2011.

View | Download