Towards a secure access control model for E-learning platform based on multi agent systems

Nowadays E-learning's popularity is increasing as more and more people are taking online courses. It becomes the need of the hour. The increase in the amount of resources available in E-learning systems will require the need for designing new architecture of such systems. The integration of multi-agent systems has played a very important role in improving the search quality, supplying new solutions and simplifying ways to very complex tasks. While a lot of efforts in the E-Iearning and multi-agent system domains has been put into delivering infrastructure and providing content, security and trust issues have hardly ever been considered. Worth knowing that in Multi-agent systems with the properties of agents (autonomous, social...), some activities will be dangerous and unreliable because it is difficult to know which agents are trustworthy and external accesses are safely. In the present paper, the security aspects are directly related to the application of a control access policy, responsible for securing interaction with agents and reinforcing it with the integration of trust. The giving work will focus on combining two access models based on RBAC model :"TrustBAC",T-SR " Dynamic RBAC with Trust-Satisfaction and Reputation for multi-agent system" . The main goal here, is to develop a new model that incorporates the advantages of both models and improve the highest degree of security in the E-Iearning platforms based on multi-agent systems.