社交网络服务中一种基于用户上下文的信任度计算方法.pdfVIP

社交网络服务中一种基于用户上下文的信任度计算方法 乔秀全 杨春 李晓峰 陈俊亮 (北京邮电大学 网络与交换技术国家重点实验室, 北京 100876) 摘 要: 电子商务中的信任问题一直备受关注. 随着近些年社交网络的流行，其真实的社交网络及信任关系开始受到关 注，基于社交网络开展各种电子商务活动逐渐成为一种新的发展趋势. 因此，研究用户间的信任度对基于社交网络的电子 商务具有重要意义. 本文通过借鉴社会心理学中人与人之间的信任产生原理，提出了社交网络中基于用户上下文的信任度 计算方法. 该方法将社交网络中用户之间的信任度分为熟悉性产生的信任度和相似性产生的信任度；同时又根据所起作用 的重要程度不同，把相似性划分为内部相似性和外部相似性，并给出了信任度计算的具体方法. 最后，仿真实验验证了该 方法的合理性和有效性. 关键词: 社交网络服务;用户上下文;人际关系;信任度 中图法分类号:TP39 文献标识码 :A A Trust Calculating Algorithm Based on Social Networking Service Users’ Context QIAO Xiu-quan YANG Chun LI Xiao-feng CHEN Jun-liang (State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, Beijing 100876, China) Abstract :Trust is a significant problem in e-commerce. As the social networking sites are becoming more and more popular in recent years, its real social network and trust relationship attract people ’s attentions. The e-commerce based on the social networking sites has become a new trend. Therefore, the research on the users’ trust for e-commerce based on SNS is very significant. By learning the trust generation principle in the psychology, this paper presented a trust calculating algorithm accroding to the SNS users’ context. This method devided the user trust among the different SNS users into two parts: one is generated by familiarity and the other one is generated by similarity. At the same time, according to the importance of the role, the similarity part was divided into internal similarity and external similarity. And also the specific trust calculation method proposed by this paper was illustrated. Finally, the simulation results