摘要
现有的可信服务选择方法将信任值表示为单一的实数,并依据信任值对候选服务进行排序,不能充分体现服务可信性的特征和用户偏好。在传统服务质量模型的基础上引入可信属性构建多维服务质量模型,并阐述信任值的计算方法。在此基础上提出一种层次分析法和偏好排序组织法结合的服务选择方法,其中,层次分析法用于计算基于用户偏好的属性权重值,偏好排序组织法用于对候选服务进行排序。实验表明该方法可以有效地表达用户对于多维服务质量属性的偏好,提高服务排序和服务选择的可信性。
In current trust-based service selection methods,the value of trust is represented as a single real number and is used as the basis on ranking the candidate services,which cannot fully reflect the features of trust and users＇ preferences.In this paper we introduce a multidimensional QoS model based on traditional QoS model,and expound the calculation of the trust.On this basis,we propose a service selection method which combines the analytic hierarchy process（AHP） with preference ranking organisation method for enrichment evaluations（PROMETHEE）,in which AHP is used to determine the weights of attributes according to users＇ preferences,and PROMETHEE is used to obtain the final rank of candidate services.Experiments show that the method can effectively express users＇ preferences for multi-QoS attributes,thus the trustworthiness of service ranking and selection is efficiently improved.
出处
《计算机应用与软件》
CSCD
北大核心
2012年第11期16-19,28,共5页
Computer Applications and Software
基金
国家自然科学基金项目(60973149)
博士点基金(2010009211022)
中科院计算机科学国家重点实验室开放基金项目(SYSKF1110)
江苏省高校科技成果产业化推进项目(JHB2011-3)
