传统基于目标函数法的模糊聚类算法是一种迭代的“爬山”算法,容易陷入局部最优解.提出了基于遗传算法与禁忌搜索结合的模糊聚类算法,综合运用遗传算法的多出发点和禁忌搜索的记忆性来改善聚类的效果,并通过迭代的遗传禁忌搜索算法产生最优聚类中心,实验中分别通过人工数据和标准数据测试验证了该算法的有效性.
Traditional fuzzy clustering algorithm based on objective function is an iterative hill-climbing algorithm and is easy to fall into local optimization. This paper puts forward the fuzzy clustering algorithm based on genetic tabu algorithm, which applies synthetically many springboards of genetic algorithm and memory property of tabu algorithm to improve the clustering effect and produces the optimal clustering center by using iterating genetic tabu search algorithm. The experiments of different datasets demonstrate that the method is an effective algorithm.