基于PageRank的有向加权复杂网络节点重要性评估方法

Evaluation Method for Node Importance in Directed-Weighted Complex Networks Based on PageRank

现有复杂网络节点重要性评估研究主要集中在无向、无权复杂网络上,未能全面客观反映真实复杂网络的情况。本文基于有向加权复杂网络模型,借鉴PageRank排名算法,并结合复杂网络节点重要性评估特点,提出节点重要性评估的新指标———DWCN-NodeRank和相应评估方法,该指标既反映出节点局部连接的特性,又从全局体现了有向加权复杂网络中整体链接关系对节点重要性的影响。采用真实的复杂网络数据集所进行的仿真实验结果表明,该方法能快速、有效地评估有向加权复杂网络节点的重要性,提高了复杂网络节点重要性评估的实用价值。

The existing evaluation methods for node importance in complex network mainly focus on undi- rected and unweighted complex networks, and can not reflect objectively the reality of real world status. Based on directed-weighted complex network model, the PageRank ranking algorithm and combined with characteristics of evaluation for node importance of the directed-weighted complex network, a new evaluation index named DWCN_ NodeRank for node importance and corresponding evaluation method are put forward. Besides the local link characteristics, the proposed index can reflect the influence of whole link relationship on node importance in directed-weighted complex networks from the global view. Simulation experiment runs on the data set of real complex networks, and the results show that the pro- posed method may quickly, effectively evaluate node importance in directed-weighted complex net- works.