一种改进的K-Means算法

An Improved K-Means Algorithm

K-Means算法是一种经典的无监督聚类分析算法,具有运行速度快易于实现等优点,但使用该算法时需要指定聚类数目,且质心的选择具有不确定性。针对上述不足,提出一种基于最大最小距离和BWP指标相结合的K-Means算法。通过对UCI数据库中的4个数据集仿真实验的结果表明,所提出的改进算法在算法的准确率、聚类效果两方面均优于传统的K-Means算法以及基于最大最小距离的K-Means算法。

K-Means algorithm is a classical unsupervised clustering analysis algorithm.Although it has the advantages of fast running speed and easy implementation,it is necessary to specify the number of clustering when using this algorithm,and the selection of centroids is uncertain.In order to verify the performance of the improved algorithm,four data sets in UCI database are simulated and the test results show that,in order to verify the performance of the improved algorithm,a K-Means algorithm based on the combination of maximum and minimum distance and BWP index is proposed in this paper.The improved algorithm proposed in this paper is superior to the traditional K-Means algorithm and the K-Means algorithm based on maximum and minimum distance in terms of accuracy and clustering effect.