融合度与K核迭代次数的节点重要性排序算法

传统的节点重要性排序算法多从单一属性角度进行分析,评价不够全面,影响排序结果的准确度。为解决这一问题,从多属性融合的角度提出一种融合度与K核迭代次数的节点重要性排序算法,从局部(即度)和全局(即K核迭代次数)两个属性对节点重要性进行综合评价,使用熵权法确定局部属性和全局属性对节点重要性的贡献权重。人工网络和真实网络的实验结果表明,该算法对节点重要性进行排序时具有较高的准确性和较好的时间效率。

Node importance ranking algorithm with fusing degree and K-shell iteration number

Only a single attribute is taken into account in most traditional node importance ranking algorithms, which leads to incomplete evaluation and affects the accuracy of ranking results. To solve this problem, a node importance ranking algorithm integrating degree value and K-shell iteration number was proposed in view of multi-attribute fusion. The importance of nodes was synthetically evaluated from perspectives of two attributes of the local (degree) and the global (K-core iteration), and the entropy weight method was applied to determine the contribution weight of local attributes and global attributes to node importance. Experimental results of artificial networks and real networks show that the proposed algorithm can accurately rank the importance of nodes and it has higher execution efficiency.