一种面向团体的影响最大化方法

发布时间：2018-04-25 23:05

本文选题：社会网络 + 影响最大化　；参考：《软件学报》2017年08期

【摘要】：影响最大化旨在从给定的社会网络中寻找出一组影响力最大的子集.现有工作大都在假设实体点(个人或博客等)影响关系已知的情况下,关注于分析单个实体点的影响力.然而在一些实际场景中,人们往往更关注区域或人群等这类团体的组合影响力,如户外广告、电视营销、疫情防控等.研究了影响力团体的选择问题:(1)基于团体的关联发现,建立了团体传播模型GIC(group independent cascade);(2)根据GIC模型,给出了贪心算法CGIM(cascade group influence maximization),搜索最具影响力的top-k团组合.在人工数据和真实数据上,实验验证了该方法的效果和效率.
[Abstract]:The aim of maximizing influence is to find a set of most influential subsets from a given social network. Most of the existing work focuses on the analysis of the impact of individual entity points under the assumption that entity points (individuals, blogs, etc.) influence relationships are known. However, in some practical situations, people tend to pay more attention to the combined influence of such groups as region or crowd, such as outdoor advertising, television marketing, epidemic prevention and control and so on. In this paper, we study the selection problem of influence groups: 1) based on the association of groups, we establish a group propagation model GIC(group independent cascade2) according to the GIC model, we give the greedy algorithm CGIM(cascade group influence maximization to search for the most influential top-k clusters. The effectiveness and efficiency of the method are verified by experiments on artificial data and real data.
【作者单位】：软件工程国家重点实验室(武汉大学);武汉大学计算机学院;武汉大学国际软件学院;云南大学信息工程学院;
【基金】：国家自然科学基金(61232002,61502347,61202033,61572376) 中央高校基本科研业务费专项资金(2042015kf00 38)~~
【分类号】：TP301.6

【相似文献】