Students' success prediction on high school entrance exam by machine learning

Technological developments affect all sectors. The education sector is no exception in this regard. The primary purpose of education is to transfer the necessary information to the students and then to measure them. Predicting the success of the student and determining a road map has a great importance for both students, their families and teachers, especially for the High School Entrance Exam (LGS), which each student can take only once. The aim of this study is to predict the LGS exam success of students using data mining methods. For this purpose, Naive Bayes, IBk (KNN), J.48, KStar, Multilayer Perceptron, Random Forest classification methods were used with the data collected from the students who took the LGS exam in the last few years. Random Forest algorithm gives the best result in estimating students' LGS exam success.