Research On Cloud Service Evaluation And Recommendation Method Based On Trust Model

In recent years,the increasingly mature cloud computing technology has become more and more widely used,and the number of users is increasing.The large-scale application makes cloud computing an important part of social production.From the commercial development of cloud computing companies to the technology.And the infrastructure of strong country cloud computing have profoundly verified that cloud computing technology is the technology trend of the Internet era.One of the characteristics of cloud computing technology is that the amount of data is huge.This is a double-edged sword for anyone in the Internet.People enjoy a certain security risk while enjoying cloud computing convenience services.When users choose cloud services,the primary consideration is security.In the cloud computing system,there is a trust relationship between cloud service providers,cloud services and users.Based on the data processing rules of D-S evidence theory,this paper quantifies the trust value between the three and establishes a relationship-aware cloud service trust model,and based on the model,a recommended method for trusted cloud services is built to provide users with reasonable and reliable cloud services to meet the security requirements of users.At present,how to quantify the trust value of cloud services,how to establish a corresponding cloud service trust model,and on this basis,recommending cloud services with high credibility to users is the key.Due to the complexity of trust relationships in cloud computing systems,users' trust evaluation of cloud services has certain uncertainty,and D-S evidence theory just has the ability to process uncertainty information.Based on the theory of convex function evidence,this paper establishes a cloud service trust model,to quantify the trust relationship between cloud service providers,cloud services and users,and uses the data processing capability of the theory to build a trusted cloud service recommended method.In the process of building a trust model,in order to quantify the trust relationship of the three,this paper defines five levels of trust.First,the trust evaluation result after the interaction between the single user and the cloud service is taken as the direct trust value;secondly,the trust evaluation result of all users,the cloud service provider trust evaluation result and the third party institution trust evaluation result are regarded as the indirect trust value;finally,the data fusion rules are used to fuse the two trust values to obtain the comprehensive trust value required by the trust model,and the comprehensive trust value is taken as the measurement standard to recommend cloud services with high credibility to users.The established trust model performs trust evaluation and recommendation on cloud services without analyzing users.In practical applications,there are also ordinary users and malicious users among users.According to the defined five trust levels,cloud service providers and cloud service providers measure the similarity of users for thirdparty organizations and classify users of the same level into one category.When a user sends a cloud service request,the cloud service recommendation can be distinguished according to the user level.In this paper,the experimental method of constructing the proposed method is verified and compared with the classical evidence theory method.The results show that the validity and robustness of the cloud service trust model based on the convex function evidence theory are guaranteed.In addition,this paper utilizes the association information between cloud services in the cloud computing system,and has a good selection effect on the highly relevant cloud service providers,and can provide reasonable cloud services according to the needs of users.