Improved Chameleon Algorithm

It is found that some parameters should be determined by hand when using Chameleon algorithm to choose K value and the threshold value of similarity degree function.It is difficult to determine these parameters without any prior domain knowledge.Aiming at this problem,this paper introduces modularity theory and proposes a clustering algorithm——M-Chameleon according to the structural equivalence similarity degree and modularity theory.Experimental results confirm that M-Chameleon can reflect the actual clustering situation objectively.

[1]  Tomaso A. Poggio,et al.  Example-Based Learning for View-Based Human Face Detection , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Yoav Freund,et al.  Experiments with a New Boosting Algorithm , 1996, ICML.

[3]  M E J Newman,et al.  Finding and evaluating community structure in networks. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[4]  Vipin Kumar,et al.  Chameleon: Hierarchical Clustering Using Dynamic Modeling , 1999, Computer.

[5]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[6]  Rafael C. González,et al.  Digital image processing using MATLAB , 2006 .

[7]  Zhengxin Chen,et al.  An iterative initial-points refinement algorithm for categorical data clustering , 2002, Pattern Recognit. Lett..

[8]  Ronald S. Burt,et al.  Positions in Networks , 1976 .

[9]  J. P. Marques de Sá,et al.  Pattern Recognition: Concepts, Methods and Applications , 2001 .

[10]  M. Newman.  Analysis of weighted networks. , 2004, Physical review. E, Statistical, nonlinear, and soft matter physics.

[11]  Narendra Ahuja,et al.  Detecting Faces in Images: A Survey , 2002, IEEE Trans. Pattern Anal. Mach. Intell..