| 密度K均值聚类算法及在复杂网络分析中的应用 |
| |
| 引用本文： | 申玉发,张晓昱,赵立强.密度K均值聚类算法及在复杂网络分析中的应用[J].河北农业技术师范学院学报,2013(4):32-36. |
| |
| 作者姓名： | 申玉发 张晓昱 赵立强 |
| |
| 作者单位： | [1]河北科技师范学院数学与信息科技学院,河北秦皇岛066004 [2]河北青县树人学校,河北秦皇岛066004 |
| |
| 摘 要： | 现代信息社会中，许多实际问题都归结为复杂网络中模块问题的研究，而聚类方法是研究复杂网络中模块性的重要方法。本研究将基于视觉原理的密度聚类算法与传统的K均值聚类算法相结合，提出了一种新的聚类算法，即密度K均值聚类算法。该算法在一定程度上克服了传统的K均值聚类算法易受异常点影响和无法确定聚类数的问题，具有对初始参数不敏感、可发现任意形状的聚类，以及能找到最优聚类等优点。基于此，以城市建通网络中交巡警指挥平台的设置问题为例，通过Matlab程序求解说明了所给出聚类算法的应用。
|
| 关 键 词： | 复杂网络 聚类分析 密度聚类算法 K均值聚类算法 |
Density-based K-means Clustering Algorithm and the Application in Complex Networks |
| |
| Authors: | SHEN Yu-fa ZHANG Xiao-yu ZHAO Li-qiang |
| |
| Institution: | 1 College of Mathematics and Information, Hebei Normal University of Science ＆ Technology, Qinhuangdao Hebei, 066004; 2 Shuren School Qingxian Hebei;China) |
| |
| Abstract: | In modem information society, many problems are co-wetted to study the modularity of complex networks. And clustering method is an important way to investigate the modularity in the field of complex networks. In this paper, a new clustering algorithm, that is density-based K-means clustering algorithm, is proposed. To some extent, the algorithm can overcome the major defects of the K-means clustering algorithm, which is easily affected by outliers and unable to determine the clustering number k. This algorithm is insensi- tive to initial parameters, and can discover clusters of arbitrary shape and can find optimal clustering. Moreover, by taking the problem of platform setting for city transportation network which can be used in transportation management command system as an example, the application of the density-based K-means clustering algorithm is illustrated. |
| |
| Keywords: | complex networks clustering analysis density-based clustering algorithm K-means algorithm |
| 本文献已被 维普 等数据库收录！ |
|
