| 一种基于密度的K-means算法 |
| |
| 引用本文： | 乔小妮,张明新,史变霞. 一种基于密度的K-means算法[J]. 电脑开发与应用, 2008, 21(10): 9-11 |
| |
| 作者姓名： | 乔小妮 张明新 史变霞 |
| |
| 作者单位： | 西北师范大学数学与信息科学学院,兰州,730070;兰州高等工业专科学校,兰州,730050 |
| |
| 摘 要： | 基于密度聚类的思想，提出了一种改进的K—means算法。算法吸取密度聚类算法的优点，利用对象的t-邻域密度作为选择初始聚类中心点的条件．选出较优的初始中心点，从而得到较好的聚类效果。通过实验表明，此方法相对于随机选取初始聚类中心点准确率较高、稳定性强、可伸缩性好。
|
| 关 键 词： | K-means算法 t-邻域密度 初始聚类中心点 |
K-means Clustering Algorithm based on Density |
| |
| Qiao Xiaoni et al. K-means Clustering Algorithm based on Density[J]. Computer Development & Applications, 2008, 21(10): 9-11 |
| |
| Authors: | Qiao Xiaoni et al |
| |
| Affiliation: | Qiao Xiaoni et al |
| |
| Abstract: | To investigate an improved k-means clustering algorithm.The method regards the traditional K-means clustering algorithm as the thinking foundation and absorbs the advantage of density based clustering.The improved method takes t-neighborhood as the condition to find better initial clustering centers.So this way can get good clusterings.According to the experiment,the improved K-means Clustering Algorithm can get higher accuracy,stronger stability and good flexibility. |
| |
| Keywords: | K-means algorithm density based clustering initial clustering centers |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
