社交网络中基于信任评估的推荐控制模型

针对社交网络中无法有效管理陌生推荐安全性难题,提出了一种基于信任的评估推荐控制模型(TRCM).该模型描述了社交网络推荐中个体角色、推荐路径、可信任值、控制规则等方面内容.模型结合社交网络个体之间取信度、信誉度评估方法,分别计算串路型、并路型和复合型3类推荐信任度,并给出了客观偏移度和信心指数评估.同时,TRCM定义了一组信任控制规则以管理社交网络中的推荐行为.实验表明,该方法是有效可行的.

Trust Evaluation based Recommendation Control Model in Social Network Site

To solve the security challenge of managing strange recommendations, a trusted recommendation control model for social network sites is proposed. TRCM describes four parts in SNS recommendation formally as: roles of individuals, recommendation routes, factors of trust and control rules. This model combines the belief among individuals and reputation to calculate three kinds of recommended trust: serial route recommendation, parallel route recommendation and overlapped route recommendation respectively. Further, factors which are named objective bias and confidence in TRCM are addressed to evaluate the reliability of recommended trust computation. Meanwhile, TRCM defines a set of trust control rules to manage the recommendation in SNS. Examinations are given to explain the efficiency and feasibility of our mechanism.