| 确定最佳聚类数的二阶差分统计法 |
| |
| 引用本文： | 崔建斌,姬安召,张科.确定最佳聚类数的二阶差分统计法[J].安徽大学学报(自然科学版),2017,41(2). |
| |
| 作者姓名： | 崔建斌 姬安召 张科 |
| |
| 作者单位： | 陇东学院数学与统计学院,甘肃庆阳,745000;陇东学院能源工程学院,甘肃庆阳,745000;西北师范大学数学与统计学院,甘肃兰州,730070 |
| |
| 基金项目： | 甘肃省科技计划基金资助项目 |
| |
| 摘 要： | 多元统计数据的聚类是分析数据的一类非常重要方法,但是如何确定最佳聚类数往往是比较困难的.针对分析多元统计数据中应用最广泛的方法之一——K均值聚类方法,提出了一种新的二阶差分统计确定最佳聚类数的方法,并将其与CH统计法、Gap统计法进行了对比.实证分析表明,论文所提方法更为简单而且有效.
|
| 关 键 词： | 二阶差分 聚类 最佳聚类数 仿真实验 |
A second-order differential statistical method for determining the optimal cluster number |
| |
| CUI Jianbin,JI Anzhao,ZHANG Ke.A second-order differential statistical method for determining the optimal cluster number[J].Journal of Anhui University(Natural Sciences),2017,41(2). |
| |
| Authors: | CUI Jianbin JI Anzhao ZHANG Ke |
| |
| Abstract: | Clustering in multivariate statistics is a very important method for data analysis.But how to determine the optimal cluster number is usually difficult.K-mean cluster is one of the most widely used method in multivariate data analysis.Based on the K-mean cluster,a new second-order differential statistical methods was proposed to determine the optimal cluster number.It was also compared with CH statistical method and Gap statistical method.Results indicated that the proposed second-order differential statistical method was more simpler and more effective. |
| |
| Keywords: | second-order differential cluster the optimal cluster number simulation experiment |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
