用于复杂网络节点重要度评估的离心率算法改进研究

复杂网络中的节点重要度评估一直备受关注。鉴于离心率中心性只考虑节点最大最短路径存在一定局限性，通过计算处理节点的平均最短路径，考虑离心率数值与平均最短路径的差值，提出改进后的新方法。在具有代表性的APAR网络上进行计算实现，并与其它节点重要性评估方法进行对比，发现该方法较离心率中心性方法，对于节点的粗略划分更加精细、有效|在SI模型的模拟对照中，发现该方法在最终第10个单位时间时，准确性相较于离心率中心性提升了15%。

Research on Improvement of Eccentricity Centrality Algorithm for Importance Evaluation of Complex Network Nodes

The evaluation of node importance in complex networks has always been of great practical significance. Since the eccentricity centrality only considers the maximum and shortest path of the node， this method has certain limitations. By calculating the average shortest path of the processing node and considering the difference between the eccentricity value and the average shortest path， we propose a new improved method， which is implemented on a representative APAR network， and compared with other node importance evaluation methods. It is found that the method improves the coarse division of nodes by the eccentricity central method more accurately and effectively. In the comparison simulation in SI model， the accuracy of the method was improved compared with the evaluation results of the eccentricity center.