一种改进禁忌搜索的K-medoids聚类算法

针对传统K-medoids聚类算法初始聚类中心随机选择、聚类精度不高、全局搜索能力较差以及禁忌搜索算法对初始值随机选取等问题,提出了一种粒计算与最大距离积法相结合的初始化禁忌搜索初始值算法,将改进后的禁忌搜索算法用来优化K-medoids,以提高聚类算法的性能。通过仿真试验论证了该算法具有较高的效率和准确率以及较强的稳定性。

K-medoids clustering algorithm based on improved tabu search

In view of the shortcomings of the traditional K-medoids clustering algorithm such as the randomly selected initial clustering center,low clustering accuracy and the poor global search ability,and the Tabu Search algorithm random selection of initial values,a new Tabu Search algorithm is proposed that the initialization of Tabu Search is based on granular computing and maximum distances product.This paper will further optimize K-medoids to improve the performance of the clustering algorithm.The experimental results show that the algorithm has higher efficiency and accuracy as well as strong stability.