A Systematic Review on Recommendation Systems Developed for the Field of Adult Education

Abstract

Today, the changing job performances and the desire of individuals to improve themselves have created the desire of people to quickly reach the content that is suitable for them. The use of suggestion systems designed to identify the needs of individuals and present the most appropriate content is considered as a solution method in this regard. The aim of this systematic study is to determine the trends in the field by making a comprehensive analysis about what kind of suggestions are given in the studies on the recommendation systems used and designed in the field of adult education, in which year the studies were conducted, the research method used, the filtering methods used and the algorithms used, and to identify the trends in this field to establish an up-to-date basis for new entrants. As a result of the review made in various databases, 113 studies were reached, and a systematic analysis of 75 studies that met the inclusion criteria was carried out. As a result of the review, it was seen that the most content suggestions were made, the most publications were made in 2020, the research method focused on determining the system performance and promoting, and a limited number of experimental studies were included. It has been determined that the most collaborative filtering method is used, and content-based and hybrid filtering methods are less preferred. It has been concluded that the K nearest neighbor algorithm is used much more than other algorithms, and besides this algorithm, artificial neural networks, support vector machines, decision trees and newly proposed algorithms by the researchers are also included. In line with the results obtained, investigations were made and suggestions were made for practice and research for future studies.