Kernel-based K-means Clustering
Li Guo · Jisuanji gongcheng · 2004
The paper applieds the idea of kernel-based learning methods to K-means clustering. It proposes an algorithm of kernel K-means clustering. The idea of the algorithm is firstly map the data from their original space to a high dimensional space (or kernel space) where the data are expected to be more separable then perform K-means clustering in the high dimensional space. Meanwhile it improves the speed of the algorithm by using a new kernel function---conditionally positive definite kernel (CPD). The performance of new algorithm is demonstrated to be superior to that of K-means clustering algorithm by experiments on artificial and real data.