社交网络中基于中心加权链接的影响力算法

摘要

针对现有解决影响最大化问题的方法局限性,考虑网络节点深层次结构对影响扩散的作用,基于中心启发式的思想,提出一种基于中心性加权链接强度的混合算法.基于线性阈值模型计算节点的潜在影响力,启发式选择周边影响力之和大于本身潜在影响力的节点作为种子节点进行激活,运用贪心算法选取具有最大影响增量的节点扩展.实验结果表明,该混合算法具有较好的激活范围以及较高精度的选择性.%In view of the existing methods,to solve the limitations of the influence maximization problem,considering the effects of network nodes deep substructure on the diffusion,a hybrid algorithm based on central idea was proposed.Based on the linear threshold model,the potential influence of nodes was calculated,the node whose surrounding influence was greater than its po-tential impact was heuristically chosen as the seed node for activation,and the greedy algorithm was used to select the node expansion with largest impact.Experimental results show that the hybrid algorithm has better activation range and higher accuracy.