Predicting Student’s Performance in Educational Institutes Using Tree Family Classifiers
Predicting Student’s Performance in Educational Institutes Using Tree Family Classifiers

Summary/Abstract: Academic systems are working in a complex environment and faced problems to analyze the performance of students using their current systems. Therefore, they use data mining tools to analyze an enormous set of data, get hidden and useful knowledge, and extract meaningful information. The aim of this study is to develop new learning techniques to predict students’ academic results. Five classifiers of the tree family used for the experiments are J48, BF Tree, NB Tree, Random Forest, Random Tree and REP Tree. The results indicated that Random Forest outperformed all the other tree family classifiers in all three datasets. Hence, it is a superior classification technique among the classifiers used for the educational datasets.

• Details
• Contents