Course Recommendation using Domain-based Cluster Knowledge and Matrix Factorization
Ishita Malhotra, Projit Chandra, R. Lavanya · 2022
Selection of elective courses is one of the most crucial choices that a student needs to make in their academic journey. This choice also helps them shape their career path and develop their skill set in a certain domain. This is also one of the choices that students mess up due to lack of knowledge about the availability of the courses, confusion among courses, and sometimes, peer influence. This research proposes a system that will help students to take up the best available elective courses according to their domains of interest. This system creates a cluster of students based on their domains of interest and uses the matrix factorization technique to look at the historical performance records of students in those domains to make predictions for the courses that can be taken by a particular student belonging to the cluster. As a result, the system is essentially filtering the best possible candidates, for a particular student, in the cluster. Using the data of those candidates, predictions were made for particular students. The proposed system can enhance the overall learning experience for students, thereby helping them develop the foundation for their careers.