A novel kernel self-organizing map algorithm for clustering

Kernel Methods are algorithms that, by replacing the inner product with an appropriate positive definite function, implicitly perform a nonlinear mapping of the input data into a high-dimensional feature space. In this paper, a novel kernel SOM (seIf-organizing map) algorithm is proposed based on energy function for solving the disadvantage lies in lack of direct descriptions about the clusterings´ centers and results in the original SOM algorithm. Furthermore, how to determine the parameters initialization is also discussed in this paper. To identify the effective of the proposed algorithm, some data are applied to test KSOM and SOM algorithm ,The result of the experiments show KSOM algorithm are good performance than SOM.