摘要
针对现有社交网络生成模型在解释网络中观结构和微观特性上的不足,提出了考虑社会选择作用的BA模型(SSBA).该模型基于社交网络用户的特征分布和同质性形成的内在机理,在建模过程中考虑了社会影响和社会选择的共同作用.在仿真数据与真实数据集上的实验表明, SSBA模型能很好地刻画不同类型社交网络在宏观特性和中观结构上的特点.发现不同作用机制下网络表现出相异的统计特性,社会选择作用较强时,网络度分布逐渐偏离幂律分布,且网络逐渐向同配网络转变.
To address the weakness of the existing generation models in explaining the mediostructure and microcosmic mechanisms of real social networks, this paper proposes a social network generation model named social selection-aware BA(SSBA). Considering the characteristic distribution of users in social networks and the internal mechanism of homogeneous, the SSBA model combines the mechanisms of social selection and social influence in the modeling process. Experiments on both simulated and real world datasets show that the SSBA can capture the macroscopic perspective and mediostructure in a variety of networks. The experimental results also indicate that networks under different mechanisms show diverse statistical characteristics: when social selection is stronger, the degree distribution gradually represents a departure from power law distribution,and the whole network gradually transforms into the an assortative network.
作者
刘业政
李玲菲
孙春华
Liu Yezheng;Li Lingfei;Sun Chunhua(School of Management,Hefei University of Technology,Hefei 230009,China)
出处
《系统工程学报》
CSCD
北大核心
2019年第5期587-597,共11页
Journal of Systems Engineering
基金
国家自然科学基金重大资助项目(71490725)
国家自然科学基金重大研究计划重点资助项目(91846201)
国家自然科学基金创新研究群体资助项目(71521001)
关键词
生成模型
社交网络
社会选择
社会影响
generation model
social networks
social selection
social influence
