摘要
构建了基于多影响因素的信任计算模型,以克服拍卖网站中现有的信任计算方法的不足以及传统理论信任模型的局限性。模型中,借鉴了现有的信任研究成果,并结合网上拍卖的特点。引入了成交价格、评分用户信任度、时间权重三项因素。参照累加信任模型,增加了信誉反馈评分、近期信任度两项因素。仿真实验比较了不同规模的恶意用户下,欺诈行为对该模型以及eB ay、Sporas模型的影响。分析和仿真说明,该模型不仅弥补了现有模型的不足,而且有较好的可行性。
A trust model was built based on multiple factors for online auctions to overcome the limitation of traditional trust model and the current reputation mechanism,considering the characteristic of C2C auctions.The model integrated 5 key factors affecting online trust together,introduced three key factors: price, trust degree of user,and time.The model presented other two key factors: feedback degree of reputation and latest trust degree.The simulation later compared the model with eBay and Sporas model,by setting variant number of malicious users while encountering fraud.The simulation analysis indicates the model is effective and feasible.
出处
《清华大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2006年第z1期1103-1108,共6页
Journal of Tsinghua University(Science and Technology)
基金
国家自然科学基金资助项目(70371004)
教育部博士点基金资助(20040006023)
关键词
信任
信任模型
在线信誉系统
trust
trust model
online reputation system