摘要
针对传统模糊C均值聚类算法(FCM)的缺陷,提出了一种基于改进遗传算法的模糊聚类方法.利用改进遗传算法强大的全局寻优能力,这种算法较好地克服了FCM算法对初始化敏感、容易陷入局部最优的缺陷.仿真实验证明,该算法具有较强的全局寻优能力和较快的收敛速度.
After analyzing the disadvatages of the fuzzy C-mean clustering algorithm, a novel Fuzzy C- mean clustering based on improved Genetic Algorithm is proposed. This algorithm not only avoids the local optima and also robust to initialization. The experimental result shows that the algorithm increases the convergence speed and has global searching capability.
出处
《西安工程大学学报》
CAS
2008年第5期605-609,共5页
Journal of Xi’an Polytechnic University
基金
陕西省教育厅自然科学专项基金资助项目(06JK286)
关键词
聚类
FCM算法
遗传算法
种群熵
clustering
fuzzy C-means
genetic algorithm
population entropy