Improving Kernel soft Subspace Clustering Algorithms using a Particle Swarm Optimization
Amel Hebboul, Fella Hachouf · 2016
In classification task, kernel functions are used to make possible to partition data that are linearly non-separable. In this paper, a Particle Swarm Optimization (PSO) is used to obtain optimal cluster centres, their weights features vectors and a kernel parameter by optimizing a cluster validity index. A comparative study has been conducted on synthetic and real dataset. The efficiency of the proposed method has been proven by the obtained results.