摘要
提出一种基于贪心随机自适应搜索过程的聚类算法.该算法先根据密度概念构造一个约束候选列表,然后从列表中随机选取k个对象作为K均值算法的k个起始中心点.试验结果表明该算法的聚类结果比k均值算法有显著改进.
This paper presents a new clustering algorithm based on a Greedy Randomized Adaptive Search Procedure（GRASP）.It firstly constructs a restrict list of candidates（RLC） according to density by greedy process.Then randomly chooses k objects from RLC as initial k center points to K-means.The proposed algorithm presented the best clustering results confirmed statistically comparing with K-means.
出处
《闽江学院学报》
2009年第5期62-65,共4页
Journal of Minjiang University
基金
福建省科技厅高校专项项目(2006J0411)
闽江学院科技启动项目(YKQ06005)
