论文部分内容阅读
现有的Web服务选择方法通常假定偏好由用户给出.由于偏好的主观性和模糊性,用户通常无法用具体数字表达清楚自己的偏好.且QoS各维属性之间存在相关性,偏好加权的方法无法消除信息的重叠,导致服务综合QoS评价不准确.针对该问题,在Web服务选择框架中对QoS属性设置区间搜索以考虑用户的优先偏好,使得初选的服务满足用户的QoS约束.对初选的服务利用主成分分析的思想,提出一种可行的Web服务选择算法PCA-WSS,根据各主成分的贡献率进行加权,分离QoS各维属性之间存在的相关性,有效地评价服务的综合QoS,为用户选择综合QoS最优的服务.实验结果验证算法的有效性和可行性.
Existing Web service selection methods usually assume that the preference is given by the user.Because of the subjectivity and fuzziness of the preference, the user can not usually express their preference in concrete numbers, and there is a correlation between the QoS dimension attributes, preference weighted This method can not eliminate the overlap of information and lead to inaccurate evaluation of comprehensive QoS of service.Aiming at this problem, we set interval search on QoS attributes in the framework of Web services selection to consider the preference of users and make the primary service meet the users’ QoS constraints. The primary service uses the principle of principal component analysis to propose a feasible Web service selection algorithm PCA-WSS, which is weighted according to the contribution rate of each principal component to separate the correlations existing in each dimension of QoS and effectively evaluate the service Comprehensive QoS for users to select the best integrated QoS services.Experimental results verify the effectiveness and feasibility of the algorithm.