Using Markov Chains and Data Mining Techniques to Predict Students’ Academic Performance

Abstract

In this study, the academic performance of students from the E-Commerce department at Palestine Technical University – Kadoorie is predicted using a Markov chains model and educational data mining. Based on the complete data regarding the achievements of the students from the 2016 cohort of students obtained from the university’s admissions and registration department, a Markov chain is built, in which the states are divided according to the semester average of the student, and the ratio of students in each state is calculated in the long run. The results obtained are compared with the data from the 2015 cohort, which demonstrates the efficiency of the Markov chains model. For educational data mining, the classification technique is applied, and the decision tree algorithm is used to predict the academic performance of the students, generalizing results with an accuracy of 41.67%.