摘要
复杂网络聚类方法可以挖掘复杂网络的结构,对复杂网络的研究具有重要意义。DBSCAN算法是一种基于密度的聚类算法,主要用于对传统数据点集进行聚类。由于复杂网络的特殊性质,对DBSCAN算法进行改进,采用相似度度量法代替传统算法中的欧式距离度量,对复杂网络进行聚类。其优点是聚类快速、可以发现任意形状的聚类、自动确定聚类数以及有效剔除噪声点。
The method of complex network clustering can excavate the structure of complex network, which is of great significance to the research of complex network.DBSCAN algorithm is a density clustering algorithm, which is used to cluster traditional data points.Due to the special nature of complex network, to improve the DBSCAN algorithm,adopt the method of similarity measure to replace the Euclidean distance measurement in the traditional DBSCAN algorithm to cluster the complex network..The advantages of this method are clustering fast, finding the clustering of arbitrary shapes, automatically determining the clustering number, and effectively eliminating the noise points.
出处
《电脑知识与技术》
2018年第1Z期141-143,共3页
Computer Knowledge and Technology
基金
北京林业大学大学生科研训练计划(项目号:X201710022145)
国家自然基金项目资助(基金号:11501032)
关键词
复杂网络
网络聚类
密度聚类
complex network
network clustering
density clustering
