Complex data clustering using a new competitive learning algorithm

This paper introduces and discusses some competitive learning algorithms for complex data clustering. A new competitive learning algorithm, named the dynamically penalized rival competitive learning algorithm (DPRCL), is introduced and studied. It is a variant of the rival penalized competitive algorithm and it performs appropriate clustering without knowing the number of clusters, by automatically driving the extra seed points far away from the input data set. It doesn't have the "dead units" problem. The results of simulations, performed in different conditions, are presented, showing that the performance of the new DPRCL algorithm is better if compared with other competitive algorithms.