Cooperative and Penalized Competitive Learning for Clustering Analysis

Competitive learning approaches with penalization or cooperation mechanism have been applied to unsupervised data clustering due to their attractive ability of automatic cluster number selection. In this paper, we further inves-tigate the properties of different competitive strategies and propose a novel learning algorithm called Cooperative and Penalized Competitive Learning (CPCL), which implements the cooperation and penalization mechanisms simultane-ously in a single competitive learning process. The integra-tion of these two different kinds of competition mechanisms enables the CPCL to have good convergence speed, preci-sion and robustness. Experiments on synthetic and real data sets are performed to investigate the proposed algorithm. The promising results demonstrate its superiority.