摘要
针对FCM算法不足,提出一种改进的模糊聚类算法:基于遗传算法(GA)与粒子群优化算法(PSO)并行的模糊聚类算法.实验结果表明,该算法比单基于GA或者PSO的模糊聚类有较好分类正确率与稳定性,有效克服了传统FCM算法对初值敏感和易陷入局部极小值的问题.
Concerning to the disadvantages of the classical FCM algorithm, this paper proposes a improved fuzzy clustering algorithm based on genetic algorithms（GA） paralleled with the particle swarm optimization （PSO）. The experiment shows that the new algorithm＇s more accurate and stable than the algorithms only using the GA or PSO; it also shows that new algorithm effectively avoids the local optima and is robust to initialization.
出处
《福州大学学报（自然科学版）》
CAS
CSCD
北大核心
2009年第3期336-339,共4页
Journal of Fuzhou University(Natural Science Edition)
