基于前驱后继节点的社会网络影响最大化算法  

Previous and Successor Nodes-Based Heuristic Algorithm for Influence Maximization

作　　者：覃俊[1] 易金莉[1] 

机构地区：[1]中南民族大学计算机科学学院,武汉430074

出　　处：《中南民族大学学报（自然科学版）》2016年第4期95-100,共6页Journal of South-Central University for Nationalities：Natural Science Edition

基　　金：国家民委基金资助项目(2015BAD29B00)

摘　　要：针对社会网络影响最大化问题,基于挖掘"潜在影响力"节点的策略并结合贪心算法可有效降低问题复杂度,综合考虑了节点与其前驱后继节点的相互影响,对"潜在影响力"进行了重新定义,基于线性阈值模型提出了基于前驱及后继节点的影响最大化算法.实验结果表明:与目前的同类算法相比,该算法具有更好的信息扩散范围.Strategies based on mining ＂potential influence＂ nodes combined with Greedy Algorithm could effectively reduce the complexity of the influence maximization problem for social networks. In this paper, we redefined ＂potential influence＂ by eonsidering the interaction influence between nodes and their previous and successor nodes, and proposed the Previous and Successor Nodes-Based Heuristic Algorithm for Influence Maximization based on linear threshold model. Experimental results demonstrate that our algorithm significantly outperforms the similar algorithms in information diffusion.

关 键 词：影响最大化 潜在影响力 前驱后继节点 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]