融入网络结构与社交习惯的不对称用户关系强度计算

伴随着大量用户内容的创建和交换,社交网络平台中产生了大规模的互动数据和复杂的用户关系,受到了越来越多研究者的关注.但是现有对关系强度研究多是从用户特征属性相似度和社交行为两方面进行,忽略了网络结构对关系强度的影响,且并未考虑社交行为存在的方向性和习惯性问题.基于此,本文提出了不对称的社交网络用户关系强度计算方法(DSTS-ATI),该方法融合用户特征属性相似度、网络结构连接强度、社交行为交互强度三个维度来综合计算用户关系.在计算网络拓扑结构连接强度时,不仅考虑了用户间邻居节点数,还考虑了邻居节点连接边数.社交互动行为发生的方向性和习惯性,会影响用户对关系强度的感知,因此本文计算出不同社交行为的贡献权重,从交互双方感知用户社交强度.实验证明,本文提出的不对称用户关系强度的方法能够提高用户关系强度预测的准确性,有助于微博意见领袖的发现和信息传播机制的研究.

A dissymmetrical tie strength estimation method based on topology of networks and habituation of social interactions

As the product of web2.0 era, the emergence of social network sites make people bridge the traditional social gap caused by the temporal and spatial distance. Its wide spread and rapid development have changed our lifestyles dramatically. With the mass creation and frequent exchange of user generated content (UGC), an amount of interactive data and complex user relationship have been generated in the social network sites. This phenomenon attracts the attention of enormous researchers. However, extant researches of the tie strength mainly focused on the user attributes and social interactions. But they ignored the influence of network structure, and did not take into account the direction and habituation of social interactions. Hence, we propose a dissymmetrical tie strength estimation method based on user attributes, topology of networks and social interactions (DSTS-ATI). From the perspective of topology of networks, we consider not only the number of common neighbor nodes, but also the links between the nodes. The direction and habituation of social interactions affect the users' perception of tie strength. Therefore, this paper gives out the weights of different social interactions bidirectionally. The experimental results show that the proposed method enhances the accuracy of tie strength prediction. Further, it is beneficial to the research of opinion leaders discovery and the information dissemination mechanism.