社会网络中基于主题的影响力最大化算法

发布时间：2018-04-28 06:55

  本文选题：社会网络 + 影响力最大化　； 参考：《计算机应用研究》2016年12期

【摘要】：为了解决现有的影响力最大化研究没有充分考虑主题对影响力节点挖掘的影响而导致特定主题下节点集合的影响范围不大这一问题,提出了一种社会网络中基于主题的影响力最大化算法TIM。该算法首先根据主题敏感阈值对初始节点集进行预处理,剔除干扰节点,再在新的节点集合上分两个阶段进行节点挖掘。第一阶段挖掘主题权威性大的节点,第二阶段挖掘主题影响增量最大的节点,最后综合两个阶段的节点作为结果集并进行实验验证。实验结果表明,相比其他算法,TIM算法挖掘的节点集合在特定主题下的影响范围更大,时间复杂度更低。
[Abstract]:In order to solve the problem that the influence of topic on impact node mining is not fully taken into account in the existing research of maximizing influence, the influence scope of node set under a specific topic is not large. A topic based influence maximization algorithm, TIMI, is proposed in this paper. The algorithm preprocesses the initial node set according to the subject-sensitive threshold, removes the interference nodes, and then mine the nodes in two stages on the new node set. In the first stage, the most authoritative nodes are mined. In the second stage, the nodes with the largest increment are mined. Finally, the nodes of the two stages are synthesized as the result set and verified by experiments. The experimental results show that compared with other algorithms, the node set mined by Tim algorithm has a larger influence range and lower time complexity under a specific topic.
【作者单位】： 江苏大学计算机科学与通信工程学院;大全集团;
【基金】：国家自然科学基金资助项目(71271117) 江苏省科技支撑计划资助项目(BE2011156)
【分类号】：TP393.09;G206
，

本文编号：1814241