一种改进的基于遗传算法的K均值聚类算法

结合遗传算法和K均值聚类算法的优点,提出一种改进的基于遗传算法的K均值聚类算法.将遗传算法的编码方法、初始化、适应度函数、选择、交叉和变异等较好地应用于聚类问题,不仅解决了K均值聚类算法中K值难以确定、对初始值敏感以及遗传算法存在收敛性差和容易早熟的缺点,而且实现了聚类中心的优化选择、K值的自动学习和基因的自适应变异等...

K-means Clustering Algorithm Based on Improved Genetic Algorithm

Based on advantages of K-means clustering algorithm and genetic algorithm,an improved genetic clustering algorithm was proposed.Encoding method,initialization,fitness function,selection,crossover and mutation of genetic algorithms were better applied to the clustering problem,which not only can solve the problems that K value is difficult to determine,sensitive to initial value in K-means clustering algorithm and that genetic algorithm has poor convergence and is easy to be precocious and other shortcomings...