A recommender systems application for students registration renewal

It could be very challenging to choose a course at the registration period for university students. A system could overcome this problem by giving academic advice. System should recommend courses to students. It is very difficult for students to find courses that match their academic profile with traditional methods like word of mouth and information in the course catalogs. In addition, even if academic advice is provided from an academician, it is not possible for the relevant academician to know the academic profile of each student. Furthermore, it is not possible for academics to know the qualifications of all courses in their departments. In this study, a course recommender system has been developed in order to increase the academic success of the students and enable them to choose the courses that fit their own profiles. The proposed recommender system is a user-based collaborative filtering recommender system. This system examines the courses previously taken by other students who has similar academic profiles with active student and provides users recommended courses list and letter grade estimates. Proposed recommender system does not require any explicit data from its users when generating this list of recommendations and letter note estimates. System creates recommendations by establishing a similarity between the performance of the active user and the upper semester students/graduates in the same department. Thus, it does not need to define the academic features and content of the courses. The system was developed by experimental methods. The results obtained from the experiments show that the system works successfully. Keywords: Recommender systems, User based collaborative filtering, Course recommendation system