An Improved Early Student’s Academic Performance Prediction Using Deep Learning

• Main Authors: Nida Muhammad Aslam, Irfan Ullah Khan, Leena H. Alamri, Ranim S. Almuslim
• Format: Article
• Language: English
• Published: Kassel University Press 2021-06-01
• Series: International Journal of Emerging Technologies in Learning (iJET)
• Subjects: deep learning; educational data mining; early prediction; smote
• Online Access: https://online-journals.org/index.php/i-jet/article/view/20699

Nowadays due to technological revolution huge amount of data is generated in every fields including education as well. Extracting the useful insights from consequential data is a very critical task. Moreover, advancement in the deep learning techniques resulted in the effective prediction and analysis of data. In our proposed study deep learning model is be used for predicting the student’s academic performance. Experiments were performed using the two courses da-ta i.e., mathematics and Portuguese course. The data set contains demograph-ic, social, educational and students course grade data. The data set suffers from the imbalance, SMOTE (synthetic minority oversampling technique) is used. We evaluate the performance of the proposed model using several fea-ture sets and evaluation measures such as precision, recall, F-score, and ac-curacy. The result showed the significance of the proposed deep learning mod-el in early prediction of the students’ academic performance. The model achieved an accuracy of 0.964 for Portuguese course data set and 0.932 using mathematics course data set. Similarly, the precision of 0.99 for Portuguese and 0.94 for mathematics.