The Fuzzy Mega-cluster: Robustifying FCM by Scaling Down Memberships

A new robust clustering scheme based on fuzzy c-means is proposed and the concept of a fuzzy mega-cluster is introduced in this paper. The fuzzy mega-cluster is conceptually similar to the noise cluster, designed to group outliers in a separate cluster. This proposed scheme, called the mega-clustering algorithm is shown to be robust against outliers. Another interesting property is its ability to distinguish between true outliers and non-outliers (vectors that are neither part of any particular cluster nor can be considered true noise). Robustness is achieved by scaling down the fuzzy memberships, as generated by FCM so that the infamous unity constraint of FCM is relaxed with the intensity of scaling differing across datum. The mega-clustering algorithm is tested on noisy data sets from literature and the results presented.

[1]  W. Peizhuang.  Pattern Recognition with Fuzzy Objective Function Algorithms (James C. Bezdek) , 1983 .

[2]  R. Davé,et al.  Noise clustering algorithm revisited , 1997, 1997 Annual Meeting of the North American Fuzzy Information Processing Society - NAFIPS (Cat. No.97TH8297).

[3]  Paul R. Kersten,et al.  Fuzzy order statistics and their application to fuzzy clustering , 1999, IEEE Trans. Fuzzy Syst..

[4]  James M. Keller,et al.  The possibilistic C-means algorithm: insights and recommendations , 1996, IEEE Trans. Fuzzy Syst..

[5]  James C. Bezdek,et al.  Pattern Recognition with Fuzzy Objective Function Algorithms , 1981, Advanced Applications in Pattern Recognition.

[6]  Wei Li-mei.  Rival Checked Fuzzy C-Means Algorithm , 2000 .

[7]  James C. Bezdek,et al.  A mixed c-means clustering model , 1997, Proceedings of 6th International Fuzzy Systems Conference.

[8]  Rajesh N. Davé,et al.  Characterization and detection of noise in clustering , 1991, Pattern Recognit. Lett..

[9]  A. Keller.  Fuzzy clustering with outliers , 2000, PeachFuzz 2000. 19th International Conference of the North American Fuzzy Information Processing Society - NAFIPS (Cat. No.00TH8500).

[10]  Xiaomin Liu,et al.  A Least Biased Fuzzy Clustering Method , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[11]  Amit Banerjee,et al.  Validating clusters using the Hopkins statistic , 2004, 2004 IEEE International Conference on Fuzzy Systems (IEEE Cat. No.04CH37542).

[12]  M. Misra,et al.  Fuzzy clustering in drug design: Application to cocaine abuse , 2004, IEEE Annual Meeting of the Fuzzy Information, 2004. Processing NAFIPS '04..

[13]  Rajesh N. Davé,et al.  Robust clustering methods: a unified view , 1997, IEEE Trans. Fuzzy Syst..

[14]  Jongwoo Kim,et al.  Application of the least trimmed squares technique to prototype-based clustering , 1996, Pattern Recognit. Lett..

[15]  Weixin Xie,et al.  Suppressed fuzzy c-means clustering algorithm , 2003, Pattern Recognit. Lett..

[16]  R. Davé,et al.  The feasible solution algorithm for fuzzy Least Trimmed Squares clustering , 2004, IEEE Annual Meeting of the Fuzzy Information, 2004. Processing NAFIPS '04..

[17]  Amit Banerjee,et al.  A Dynamic Game-theoretic Approach to the Modified Prisoner's Dilemma , 2005, Log. J. IGPL.

[18]  James M. Keller,et al.  A possibilistic approach to clustering , 1993, IEEE Trans. Fuzzy Syst..

[19]  Moshe Kam,et al.  A noise-resistant fuzzy c means algorithm for clustering , 1998, 1998 IEEE International Conference on Fuzzy Systems Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36228).