Louvain算法与K均值聚类算法的比较研究
A Comparative Study of Louvain Algorithm and K-Means Clustering Algorithm
-
摘要: 复杂网络是近年来新兴的研究领域,社区发现是其应用方向之一.对于现实数据集进行聚类分析是数据挖掘的一个重要方法,但存在聚类分析效果不佳的情形. 此时若引入相关性度量,将数据集构建成复杂网络, 便可使用社区发现方法对其进行处理.现有文献大多针对算法进行改进, 对两种方法的划分结果进行比较的研究较少.本文选取了社团划分中的Louvain算法与聚类算法中的K均值聚类算法,首先对两种算法的理论进行比较, 接着利用心脏病、肾病患者数据构造复杂网络,比较了Louvain算法的社区划分结果与K均值聚类算法的聚类结果,在正确划分率的评价标准下,Louvain算法的社团划分结果优于K均值聚类算法的聚类结果.Abstract: Complex network is a new research field in recent years, and community discovery is one of its application directions. Clustering analysis of real data sets is an important method of data mining, but there is a situation that the effect of clustering analysis is not good. At this time, if the correlation measure is introduced to construct the data set into a complex network, the community discovery method can be used to process it. Most of the existing literatures have improved the algorithm, and few studies have compared the results of the two methods. In this paper, the Louvain algorithm in community division and the K-means clustering algorithmin clustering algorithm are selected. First, the theories of the two algorithms are compared. Then, the complex network is constructed by using the data of patients with heart disease and kidney disease. The results of the community division of the Louvain algorithm and the clustering results of the K-means clustering algorithm are compared. Under the evaluation criteria of the correct division rate, the results of the community division of the Louvain algorithm are better than the clustering results of the K-means clustering algorithm.