Shooting SOM and its Application for Clustering

Abstract—The Self-Organizing Map (SOM) is popular algorithm for unsupervised learning and vi-sualization introduced by Teuvo Kohonen. One of the most attractive applications of SOM is clustering and several algorithms for various kinds of clustering problems have been reported and investigated. In this study, we propose a new type of SOM algorithm, which is called Shooting SOM (SSOM) algorithm. The im-portant feature of SSOM is that the neurons move like aiming at a target, namely, some neurons find near the center of the area where input data are concentrated and other neurons get away from the former neurons. Because of this feature, SSOM tends to self-organize only in the area where input data are concentrated. We investigate the behavior of SSOM and apply SSOM to clustering problems. 1