Kasra Madadipouya

Mohammad Ahmadi, Mostafa Vali, Farez Moghaddam et al.

Security issues are the most challenging problems in cloud computing environments as an emerging technology. Regarding to this importance, an efficient and reliable user authentication and data protection model has been presented in this paper to increase the rate of reliability cloud-based environments. Accordingly, two encryption procedures have been established in an independent middleware (Agent) to perform the process of user authentication, access control, and data protection in cloud servers. AES has been used as a symmetric cryptography algorithm in cloud servers and RSA has been used as an asymmetric cryptography algorithm in Agent servers. The theoretical evaluation of the proposed model shows that the ability of resistance in face with possible attacks and unpredictable events has been enhanced considerably in comparison with similar models because of using dual encryption and an independent middleware during user authentication and data protection procedures.

Kasra Madadipouya

Available recommender systems mostly provide recommendations based on the users preferences by utilizing traditional methods such as collaborative filtering which only relies on the similarities between users and items. However, collaborative filtering might lead to provide poor recommendation because it does not rely on other useful available data such as users locations and hence the accuracy of the recommendations could be very low and inefficient. This could be very obvious in the systems that locations would affect users preferences highly such as movie recommender systems. In this paper a new location-based movie recommender system based on the collaborative filtering is introduced for enhancing the accuracy and the quality of recommendations. In this approach, users locations have been utilized and take in consideration in the entire processing of the recommendations and peer selections. The potential of the proposed approach in providing novel and better quality recommendations have been discussed through experiments in real datasets.