Exploring Research Supervisor Recommendation: Student Centric or Supervisor Centric

Securing an appropriate supervisor is essential for a student's research trajectory in the education field. Supervisors can provide students with dynamic and ongoing information. A supervisor recommendation system is a methodology for assisting students in selecting the most suited supervisor. This paper aims to provide an overview of research supervisor recommender systems by reviewing published papers and examining the difficulties and prospective research paths concerning this subject. After conducting a thorough search and utilizing our available knowledge, we believe this is the first review article on the Research Supervisor Recommendation System. This investigation explores the basic principles of recommendation systems and the various methods utilized to design supervisor recommendation systems. To avoid information overload, a supervisor recommendation system derives user preferences or interests from connected data sets. In this paper, we analyzed nine studies and classified them based on the preference given while recommending, i.e., student-centric where preference is given to students, supervisor-centric where preference is given to the supervisor or hybrid approaches where we evaluate data from both the supervisor and student. Three types of supervisor recommendation algorithms have been identified: content-based filtering, collaborative filtering, and hybrid filtering. In addition, this study examines the various algorithms used in each type of recommendation system.