A novel random fuzziness clustering with entropy criterion
Nianyun Shi, Liang Yan, Jiuyun Xu, Youxiang Duan · 2009
As a newly-proposed clustering algorithm based on random fuzziness model, RFKM has improved performance compared with other fuzzy clustering algorithms. However the low mobility of accuracy will lead to local optimal solution. To solve this problem, we present an Entropy-based FRKM (ERFKM) algorithm. Meanwhile, in order better to facilitate the optimal operation of the ERFKM, this paper applies entropy onto ERFKM to choose a near optimal dominant set. Simulation study indicates that the proposed is effective in clustering and it has improved performance with respect to the original RFKM.