基于介度熵的复杂网络节点重要度识别方法

现有复杂网络通常会受到随机攻击和蓄意攻击,导致复杂网络拓扑结构的可靠性性能下降.为了解决以上问题,基于图熵,结合节点的介数中心性和其所有邻居节点的度中心性,提出一种新的攻击策略,即介度熵( BE),用来识别网络中的重要节点并加以保护.实验分别通过静态攻击和动态攻击来评估攻击策略在3个标准网络模型和3个真实网络上的效率,通过比较,介度熵比传统的攻击策略具有更高的攻击效率.

Importance identification method of complex network nodes based on betweenness and degree entropy

Existent complex networks will usually be subject to random attacks and deliberate attacks, resulting in reduced reliability of complex network topologies. In order to solve the above-mentione problems, a new attack strategy is proposed based on graph entropy and incorporated with the betweenness centrality of nodes and the degree centrality of all their neighbor nodes, namely the betweenness and degree entropy (BE), which is used to identify important nodes in the network and protect them. Experimentally, the efficiency of attack strategies of three standard network models and three real networks is evaluated under condition of static attacks and dynamic attacks. By means of comparison, it is found that the betweenness and degree entropy will have higher attack efficiency than the traditional attack strategy.