doi: 10.7763/IJET.2011.V3.241
DOFCM: A Robust Clustering Technique Based upon Density
Kaur Prabhjot
,
Lamba I. M. S
,
Gosain Anjana
- Department of Information Technology at Maharaja Surajmal Institute of Technology , C-4, Janakpuri, New Delhi,110058, India . IEEE member ID- 90526150
Abstract
Robust clustering methods reduce the impact of outliers on cluster centroids. Definition of outlier depends on the data structure and applied detection methods. Noise Clustering (NC) is a robust technique, which defines outlier in terms of a distance, called noise distance. NC identifies outliers during clustering process and modifies various parameters, required for creating clusters, thus effecting clustering output. Its main motive is to reduce the influence of outliers on cluster centroids rather than identifying it hence could not result into original clusters. However, in many applications, identification of outliers is important, as they may contain important information. Density Oriented Fuzzy C - Means (DOFCM) is a robust technique, which identifies outlier before clustering, on the basis of density of data-set. According to DOFCM, outliers are defined as the points that are not in the dense part of the data-set. In this paper, we have compared both the techniques for outlier identification and clustering. The results obtained through comparison, by implementing various tests, concluded that DOFCM based upon density approach identifies outliers very well and gives efficient clustering results than NC technique which identify outliers based upon distance.
Keywords
- Data mining
- Density-Oriented approach
- Fuzzy clustering
- Outlier identification
- Robust clustering
How to Cite
Kaur Prabhjot, Lamba I. M. S, and Gosain Anjana, "DOFCM: A Robust Clustering Technique Based upon Density," International Journal of Engineering and Technology, vol. 3, no. 3, pp. 297-303, 2011. https://doi.org/ 10.7763/IJET.2011.V3.241
Copyright & License
Copyright © 2011 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).