Community self-organizing map and its application to data extraction

The self-organizing map (SOM) is a famous algorithm for the unsupervised learning and visualization 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. This study proposes the community self-organizing map (CSOM) algorithm which reflects the community in the human society. In CSOM algorithm, the neurons create some communities according to their winning frequency. We apply CSOM to various input data for clustering and data extraction, and we investigate its behaviors. We confirm that CSOM creates some communities and obtain efficient results for data extraction.