Омер Фарук АкмешеХакан КьорХасан Ербей2024-05-242024-05-242021-04-2510.33407/itlt.v82i2.4178https://acikarsiv.thk.edu.tr/handle/123456789/283<jats:p>The machine learning method, which is a sub-branch of artificial intelligence and which makes predictions with mathematical and statistical operations, is used frequently in education as in every field of life. Nowadays, it is seen that millions of data are recorded continuously, and a large amount of data accumulation has occurred. Although data accumulation increases exponentially, the number of analysts and their capabilities to process these data are insufficient. Although we live in the information age, it is more accurate to say that we live in the data age. By using stored and accumulated data, it is becoming increasingly essential to reveal meaningful relationships and trends and to make predictions for the future. It is important to analyze the data obtained from the education process and to evaluate the success of the students and the factors affecting success. These analyses may also contribute to future training activities. In this study, a data set, including socio-demographic variables of students enrolled in distance education at Hitit University, was used. The authors estimated the success of the students with demographic and social variables such as age, gender, city, family income, family education level. The primary purpose is to provide students with information about their estimated academic achievement at the beginning of the process. Thus, at the beginning of the education process, students' success can be increased by informing the students who are predicted to be unsuccessful. Diversification and enhancement of this data may also support other decision-making mechanisms in the training process. Additionally, the factors affecting students’ academic success were researched, and the students' educational outcomes were evaluated. Prediction success was compared using various machine learning algorithms. As a result of the analysis, it was determined that the Random Forest algorithm was more predictive of student achievement than others.</jats:p>journal-article