In these days, with the rapid development of information technologies and the education system of higher educational institutions, huge amounts of data are accumulating, and a large number of available courses are being developed. Consequently, students face difficulties in finding suitable courses that match their interests. As a solution, several course recommendation systems have been developed over the course of a decade, and many data mining methods for cluster data have been applied. The recommendation system allows students to notice their preferences and returns results that are useful to them, based on the assessments of other users and the assumptions of the system itself. With the help of recommendation systems, the student's learning process will be planned more productively and efficiently. The purpose of this study is to determine the general criteria of the recommendation system to meet the interests and objectives of students. In order to gain a deep theoretical understanding, a thorough review of the literature on works published over a 5-year period (2015-2020) was conducted. The paper analyzes the technologies that are used to create recommendation systems. The results obtained show common approaches, algorithms, and evaluation measurements of the recommendation system.