Forecasting students’ performance using an ensemble SSL algorithm Print

I.E. Livieris, V. Tampakas, N. Kyriakidou, T. Mikropoulos and P. Pintelas. Forecasting students’ performance using an ensemble SSL algorithm. In IEEE 1st International Conference on Technology and Innovation in Learning, Teaching and Education, 2018.



Abstract - Educational data mining is a growing academic research area which aims to gain significant insights on student behavior, interactions
and performance by applying data mining methods on educational data. During the last decades, a variety of accurate models has been developed to monitor students' future progress, while most of these studies are based on supervised classification methods. In this work, we propose an ensemble semi-supervised algorithm for the prediction of students' performance in the final examinations at the end of academic year. The experimental results demonstrate the efficiency and robustness of the proposed algorithm compared to some classical classification algorithms, in terms of accuracy.