Automatic Clustering Based on Cluster Nearest Neighbor Distance (CNND) Algorithm

This article describes a simple and fast algorithm that can automatically detect any number of well separated clusters, which may be of any shape e.g. convex and/or non-convex. This is in contrast to most of the existing clustering algorithms that assume a value for the number of clusters and/or a particular cluster structure. This algorithm is based on the principle that there is a definite threshold in the intra-cluster distances between nearest neighbors in the same cluster. Promising results on both real and artificial datasets have been included to show the effectiveness of the proposed technique.

[1]  James C. Bezdek,et al.  Efficient Implementation of the Fuzzy c-Means Clustering Algorithms , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Jian Jhen Chen,et al.  K-means clustering versus validation measures: a data-distribution perspective. , 2009, IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society.

[3]  Sanghamitra Bandyopadhyay,et al.  A symmetry based multiobjective clustering technique for automatic evolution of clusters , 2010, Pattern Recognit..

[4]  Anil K. Jain,et al.  Algorithms for Clustering Data , 1988 .

[5]  I. Guyon,et al.  Detecting stable clusters using principal component analysis. , 2003, Methods in molecular biology.

[6]  Anil K. Jain,et al.  Data clustering: a review , 1999, CSUR.

[7]  Anil K. Jain Data clustering: 50 years beyond K-means , 2010, Pattern Recognit. Lett..