Application of fuzzy-based pattern recognition techniques for cluster finding in a preshower detector in high energy heavy ion experiments

We apply the Fuzzy c-mean clustering algorithm for the data set consisting of hits in a highly granular preshower detector in the case of collision of heavy nuclei at high energy. A set of validity indexes are examined for optimum clustering performances suitable for the particular data set. The performance is also studied for different densities of hit pattern. Using information from one of the validity indexes giving good performance, FCM clustering is also studied for fixed set of parameters. The results compare well with those obtained using iterative optimization.