A Multi-class Support Vector Machine Approach for Students Academic Performance Prediction

To date, students’ educational data is still one of the most importance resources in institutions of higher learning. One way to achieving qualitative education standard by institutions of higher learning is to properly evaluate and predict the performance of entrant students and suggest appropriate faculty programmes for them based on their educational data. Several attempts have been made to predict the performance of students and placement into appropriate faculty programmes prior to admission without much success. In an attempt to correctly predict students’ performance in order to admit them into appropriate faculty programmes, a multi-class support vector machine (MSVM) predictor was built in this study. The performance of the MSVM predictor was examined using educational dataset of students from the University of Lagos, Nigeria. Findings from our experiment show that the MSVM with K-fold (K=7) cross validation adequately predicted the performances of students across all categories.