Research On User Trust Evaluation Model And Algorithm In The E-learning Services

With the rapid development of information technology and the growth of the modern educationneeds, E-learning (Electronic learning) has gradually become the important way of learning inpeople’s life. E-learning can transfer information and share learning resources without limited by timeand space because it has good interactivity and smaller time and space constraints, E-learning hasgradually evolved into services and E-learning services have been widely used in basic education andcorporate training. E-learning services not only bring the learning conveniences to people, but alsobring some unknown risks to people. The intrusion of shammers and the malicious behavior of userswill threaten the safety of services and the quality of services implementation. Therefore, it isimportant to research on user trust evaluation model and algorithm for services security.Considering the subjectivity of user itself and the objectivity of the user behavior capability, wecarry on user trust evaluation research in this paper. The main research results are summarized asfollows:1. User trust cloud assessment model has been proposed based on the user’s subjectivity.The user trust is subjective, uncertain and fuzzy. User’s expectation, entropy and hyper entropy aredefined according to complex nature of user trust. The user trust cloud evaluation model is proposedto assess the comprehensive subjective trust of user, by combing the expectation, entropy and hyperentropy.2. User’s behavior capability trust evaluation model has been proposed based on the objectivity ofuser capability.User capability trust is objective, which is reflected by user behavior interaction activities. Theuser’s behavior capability trust evaluation model is presented according to interaction activities with npeople based on m behavior capability properties; the trust assessment value is calculated by behaviorinteraction matrix operations. At the same time, the weights allocation based on informationinteraction is implemented in this model.3. User comprehensive trust evaluation model has been proposed.The user comprehensive trust evaluation model is put forward based on user trust cloud evaluationmodel and user behavior capability trust evaluation model, which the weights are allocated based onthe user expectation. Using the model we can assess user comprehensive trust effectively.4. The framework for user trust evaluation system is established. The main modules of the trustevaluation system and trust assessment update mechanism are designed and user trust evaluationsystem architecture has been implemented. Finally, the simulation results verify the feasibility andeffectiveness of the proposed model.