Trust Computation Framework based on User Behavior and Recommendation in Cloud Computing

摘要

Cloud computing provides shared environment for different resources and services that are available for users at anytime and from anywhere. Cloud computing has gained considerable attention of users and businesses. However, security concern is one of the major hurdles for acceptance of cloud computing. In order to guarantee security of data, it is necessary to grant access of data, only to authorized users. The traditional system applies different access policies and permission while granting access to any user. The analysis of user behavior is also important aspect, which can be integrated into access control model. In this paper, the trust computation model is presented that takes user behavior into consideration while providing access to the cloud data. The recommendation for the user is also one of the important components to assess user behavior. The proposed model evaluates trustworthiness of user on basis of reputation and recommendation. With the advent in machine learning techniques, applying learning based techniques in security domain has gained lots of popularity. In the proposed method, the machine learning technique (k-means clustering Algorithm) is incorporated in the trust computation process and the users are classified according their trust values.