摘要
针对动态社交网络中的社区检测问题,提出一种面向社交网络重要信息传播的重叠节点挖掘模型(SNONMM),结合标签传播算法(LPA)和扩散激活原理,实现对动态社交网络中重叠社区的高效检测.该模型的新节点在社交网络中向其他节点传播其标签的机会大于旧节点,从而使新节点更容易被发现并纳入相应的社区.同时,引入激活值来表示每个标签的传播强度,有助于更准确地捕捉社区结构的变化.为了验证该方法的有效性,通过两个真实数据集和一个人工合成网络对其性能进行评估.实验结果表明,该方法在检测社区准确性方面优于其他可用方法.
This paper addresses the problem of community detection in dynamic social networks and proposes a Social Network Overlapping Node Mining Model(SNONMM),and aims at the efficient detection of overlapping communities in dynamic social networks.The model combines the Label Propagation Algorithm(LPA)and the Spreading Activation principle to achieve efficient detection of overlapping communities in dynamic social networks.In this approach,new nodes have a greater chance of spreading their labels to other nodes in the social network than that of old nodes,making new nodes more easily discovered and incorporated into their respective communities.At the same time,activation values are introduced to represent the propagation strength of each label,which helps to more accurately capture changes in community structure.To validate the effectiveness of the proposed method,its performance was evaluated using two real-world datasets and a synthetic network.Experimental results demonstrate that the proposed method outperforms other available methods in terms of community detection accuracy.
作者
魏会廷
陈永光
WEI Huiting;CHEN Yongguang(Laboratory and Equipment Management Center,Xuchang University,Xuchang Henan 461000,China;School of Education Science,Zhoukou Normal University,Zhoukou Henan 466001,China)
出处
《西南大学学报(自然科学版)》
CSCD
北大核心
2024年第2期150-158,共9页
Journal of Southwest University(Natural Science Edition)
基金
国家自然科学基金项目(120001471)
河南省2023年哲学社会科学规划项目(2023BJY041)。
关键词
动态社交网络
社区检测
标签传播算法
扩散激活
dynamic social networks
community detection
label propagation algorithm
diffusion activation