基于随机抽样的加速 K-均值聚类方法

针对传统K-均值聚类方法不能有效处理大规模数据聚类的问题，提出一种基于随机抽样的加速K-均值聚类（K-means Clustering Algorithm Based on Random Sampling ， Kmeans＿RS）方法，以提高传统K-均值聚类方法的效率。首先从大规模的聚类数据集中进行随机抽样，得到规模较小的工作集，在工作集上进行传统K-均值聚类，得到聚类中心和半径，并得到抽样结果；然后通过衡量剩下的聚类样本与已得到的抽样结果之间的关系，对剩余的样本进行归类。该方法通过随机抽样大大地减小了参与K-均值聚类的问题规模，从而有效提高了聚类效率，可解决大规模数据的聚类问题。实验结果表明，Kmeans＿RS方法在大规模数据集中在保持聚类效果的同时大幅度提高了聚类效率。

A Speeding K-means Clustering Method Based on S ampling

To solve problems that traditional K -means clustering algorithm can not solve the large scale dataset clustering , this pa-per presents a speeding K-means clustering method based on random sampling ,called Kmeans_RS clustering algorithm .The working set is selected from the original clustering dataset by random sampling and the traditional K -means clustering method is executed on this working set .Then the center and radius of every cluster is computed and the sampling result is obtained .The last clustering result of all dataset is obtained by measuring the relationship of sampling result and other data to cluster the remai -ning data.The random sampling way is used in this process and the size of K-means clustering is decreased , so the clustering ef-ficiency is improved largely and it can be used to solve the large scale clustering problems .Simulation results demonstrate that the excellent clustering efficiency is obtained by this parallel speeding K-means method .