Towards Prediction of Students’ Academic Performance in Secondary School Using Decision Trees

Abstract

Abstract - Prediction of students’ academic performance with high accuracy is useful in many contexts. Institutions would like to know which students are likely to have low academic achievements or need assistance in order to finish their studies. Various machine learning techniques have been applied to create models to predict student’s academic performance at various levels of study. This paper aimed to develop a machine learning model for prediction of secondary school students’ academic performance. We collected records of 1720 former secondary school graduates from five public institutions in Kenya. Prediction was done by applying J48 Decision Tree, Naïve Bayes and Neural Networks Multilayer Perceptron classification techniques using WEKA machine learning environment. The study found out that J48 Decision Tree classifier predicted students’ academic performance with higher accuracy than Naïve Bayes and Neural Networks classifiers. This knowledge will help educational institutions to accurately predict academic performance of the students.