Student Graduation Prediction with Data Mining: Amasya University Distance Education Application and Research Center Sample

Correct orientation of students during education is very crucial for preventing future failures. Students who cannot graduate at the time may cause negative impacts on the family and the country's economy as well as the decrease of the young labor force. This situation requires the studies to be held concerning the students who cannot graduate in time. Analyzing the educational data related to the students is included in this study. Considering the accumulation of large number of educational data in higher education institutions, it becomes more important to analyze these data with various methods. In data mining, which is one of the methods used to analyze the data, estimation, classification and clustering methods are benefited. In this study, inferences were tried to be made about whether the students who enrolled to Amasya University Distance Education Child Development, Medical Documentation and Secretariat Associated Degree Programs in 2016-2017 will be able to graduate on time or not. Algorithms such as Decision Tree, Naïve Bayes, Support Vector Machines and Random Forest have been used for estimation. The best estimation success was reached with the decision tree algorithm. The comparative study of the estimation classification of algorithms is included in the study