摘要
模糊C-均值聚类(FCM)对初始值很敏感,易于陷入局部极小点而不能搜索到全局的聚类中心,遗传算法是一种通过模拟自然进化过程的搜索最优解的方法。因此,将FCM算法引入遗传算法的进化中,代替原来的交叉操作。实验结果表明,新方法明显优于传统FCM算法。
Fuzzy C-Means cluster algorithm usually leads to local minimum,its shortcoming is the sensibility to initial value.Genetic algorithm is a computational model of the human evolution.The application of FCM algorithm is introduced in the evolution of genetic algorithm,instead of the original crossover operator.The result shows that the new algorithm is superior to the traditional FCM algorithm.
出处
《科学技术与工程》
2010年第28期7037-7039,共3页
Science Technology and Engineering
关键词
聚类
模糊C-均值算法
遗传算法
clustering fuzzy C-means cluster algorithm genetic algorithm