摘要
符号网络是一种具有正负项关系的社会网络,对其社区结构的检测可以挖掘其中有用的信息。将符号网络 划分为全正和全负的子矩阵分别进行操作,同时引入更加适合符号网络社区检测的相关策略,如基于标签传播的 种群初始化、改进的双点交叉算子、带局部搜索的突变算子等。在基准网络和随机网络上的测试数据表明,本文 算法具有较好的检测社区检测效果。
Signed network is a kind of social network with positive and negative relationship.The detection of its community structure can mine useful information.In this paper,the signed network is divided into fully positive and fully negative sub-matrices to operate separately,and at the same time,relevant strategies that are more suitable for symbolic network community detection are introduced,such as population initialization based on label propagation,improved two-point crossover operator,mutation with local search operators etc.The test data on the benchmark network and the random network show that the algorithm in this paper has a better detection effect in the detection community.
作者
谭玉玲
TAN Yu-ling(Department of Information Engineering,Luoding Polytechnic,Guangdong Luoding 527200,China)
出处
《齐齐哈尔大学学报（自然科学版）》
2021年第6期1-8,共8页
Journal of Qiqihar University(Natural Science Edition)
基金
广东省高职高专云计算与大数据专业委员会2019年度课题(GDYJSKT19-05)
教育部科技发展中心“天诚汇智”创新促教基金课题(2018E01020)。
关键词
复杂网络
符号网络
社区检测
进化算法
complex network
signed network
community detection
evolutionary algorithm
