一种基于信度函数的复杂网络重要节点识别方法

A New Method to Identify Influential Nodes in Complex Networks Based on Belief Function

识别重要节点是复杂网络研究领域的热门话题之一,在网络的管理、维护和优化等领域有着广泛的应用价值。为了提高识别结果的准确性与全面性,本文提出一种基于信度函数的节点重要度识别方法,将节点的度、紧密度和结构洞等属性视为不同的信度函数,并运用Dempster组合规则进行融合,最终得到节点的综合评价指标值。新提出的方法不仅考虑了节点在网络中的位置,还考虑了节点的邻居数量及邻居的影响力等。实例验证表明,该方法能综合利用单一算法的优点,克服单一算法的不足,具有可行性。

Identification of influential nodes is one of the hottest research issues in complex networks,and it can be widely used in the fields of management,maintenance and optimization of the structure of networks. In order to obtain more precise and comprehensive results,a new method based on belief function to identify node importance is proposed in the manuscript. The attributes of degree,closeness and structure holes of nodes are regarded as different belief functions,and the combination rule of Dempster is carried out to aggregate them. After that,a comprehensive assessment index of node will be obtained. The factors of location in the network,the number of neighbor and the influence of neighbors of node are all taken into consideration in the new proposed model. A numerical example demonstrates that the new proposed method is of effectiveness,which integrates the advantages and overcomes the disadvantages of single method.