论文部分内容阅读
在聚类分析中,决定聚类质量的关键是确定最佳聚类数.对此,从样本几何结构的角度定义了样本聚类距离和样本聚类离差距离,设计了一种新的聚类有效性指标.在此基础上,提出一种基于近邻传播算法确定样本最佳聚类数的方法.理论研究和实验结果表明,所提出的指标和方法能够有效地对聚类结果进行评估,适合于确定样本的最佳聚类数.
In the cluster analysis, the key to determine the quality of the cluster is to determine the optimal number of clusters. In this paper, the distance between the sample cluster and the sample cluster is defined from the perspective of the sample geometry. A new poly Based on which, a method based on nearest neighbor propagation algorithm is proposed to determine the optimal clustering number of samples.Theoretical and experimental results show that the proposed index and method can effectively evaluate the clustering results, The best number of clusters suitable for determining the sample.