Research On Service Selection Algorithm Based On Trust

The development of the putting forward of service oriented calculation(SOC) and achieving of the logic structure based of SOC by service oriented architecture(SOA) have provided the latest development direction for the current distributed computing. The development of the related technology is very quickly in the new direction. Such as Web service is the mainstream technology of the SOC/SOA, and the characteristics of grid service is introducing of the SOC and Web service technology, and the pervasive computing , the semantic Web have close connection in this field.―Service‖which is the basic computing element of the distributed system is widely used in the field of distributed computing, The method of choosing effective service has become the research focus. It is easy to provide by malicious changes that increase uncertainty and service resources can not be controlled because the resources are multifarious, usually belong to different autonomous domains and are lack of a professional, trusted third–party certification entity to its certification service, meanwhile the process of implementation is not visible to the users. For these reasons, the traditional QoS-based service selection algorithm enhances the quality of service options, but it is not a good solution to the issue of trust. So how to establish appropriate service trust model and service selection algorithm based on the trust model have become a research hotspot.To solve the above problem, this paper will research from the trust model and service selection of component service and service workflow based on the trust model. Main content and contribution are as follows:(1) Learning the method of calculating similarity of recommended system, we measure the position of the recommended users by the similarity. The recommended trust model is established by the position of the recommended users and the transfer of trust properties. (2) Current trust evaluation approaches solely based on fuzzy logic and feedback are inaccurate and ineffective, this paper presents a method which bases on Set Pair Analysis theory to overcome the shortcoming that a fuzzy concept must be denoted by a precise, the only strict membership function in fuzzy math. This model considers not only the feedback information, but also the service capacity that resources could achieve. The authors propose two algorithms of service selection based on the trust model.(3) Studying the service selection algorithm of workflow based on trust value-add, this paper designs a discrete quantum particle swarm algorithm for enhancing trust. Comparing with traditional service selection algorithm, which is QoS-oriented for global optimization, this algorithm considers the issue of trust enhanced which can solve the issue of malicious deception, meanwhile this algorithm disperses quantum particle swarm combines with the features of workflow. It redefines the computed methods of various locations and the auto-regulation of weight coefficient in quantum particle swarm algorithm according to the scenarios of service selection.