AI-Based Student Dropout Prediction System

Student dropout is a serious issue for higher education institutions. From the studies reviewed, it can be seen that Artificial Intelligence and also the Machine Learning can help in finding students who may be at risk of leaving their studies. Information such as marks, attendance, previous academic records, and student behavior can be useful for making these predictions.   Different ML methods, such as Decision Tree, Random Forest, Logistic Regression, SVM, and Artificial Neural Network, have been used for finding students who may be at risk of leaving their studies. Student performance can be checked using measures such as the accuracy of the student attendance, precision, recall, F1-score, and AUC. Explainable AI methods such as SHAP and LIME can also help in understanding the reasons behind a prediction. However, issues such as missing data of student, class imbalance, fairness, and differences between institutions still need attention. Also, predicting a student at risk does not by itself prevent dropout. The prediction should help teachers and institutions provide suitable support at the right time.   Thus, AI-based student dropout prediction can be useful as an early-warning approach. Future studies should focus on improving prediction reliability, making the results easier to understand, and connecting predictions with effective student-support activities.

[1]         Rodrigues, H., Santiago, E., Wanderley, G., et al.: Artificial Intelligence Algorithms to Predict College Students' Dropout: A Systematic Mapping Study. In: International Conference on Agents and Artificial Intelligence (2024). doi:10.5220/0012348000003636  [2]    Silva, J., Roman, N.T.: Predicting Dropout in Higher Education: A Systematic Review. In: Anais do XXXII Simpósio Brasileiro de Informática na Educação (2021). doi:10.5753/sbie.2021.217437 [3]     Girón, D.A., Rosas, J.S., Marín-Rodriguez, W., et al.: Predicting Student Dropout Based on Machine Learning and Deep Learning: A Systematic Review. EAI Endorsed Transactions on Scalable Information Systems (2023). doi:10.4108/eetsis.3586 [4]          Agrusti, F., Mezzini, M., Bonavolontà, G.: Deep Learning Approach for Predicting University Dropout: A Case Study at Roma Tre University (2020). doi:10.20368/1971-8829/1135192 [5]        Sultan, M., Elsayed, H., Khan, M., Abduljabar, M.: A Deep Learning Model for MOOC Dropout Prediction Using Learner's Course-Relevant Activities. In: Global Conference on Artificial Intelligence and Internet of Things (2022). doi:10.1109/GCAIoT57150.2022.10019063 [6]    Anjali, C., Bai, V.R.: An Early Prediction of Dropouts for At-Risk Scholars in MOOCs Using Deep Learning. In: 2022 International Conference on Computing, Communication, Security and Intelligent Systems (2022). doi:10.1109/IC3SIS54991.2022.9885328 [7]      Pathiranagama, M., Nanayakkara, I., Abeyweera, R., Halloluwa, T.: Beyond the Black Box: An Interpretable Machine Learning Approach to Student Dropout Prediction. In: International Conference on                 Automation   and     Computing(2025). doi:10.1109/ICAC69156.2025.11361534  [8]         Pérez, M., Navarrete, D., Calisto, M.G.B., et al.: Unlocking Student Success: Applying Machine Learning for Predicting Student Dropout in Higher Education. In: International Symposium on Digital Forensics and   Security (2025). doi:10.1109/ISDFS65363.2025.11012013 [9]     Goren, O., Cohen, L., Rubinstein, A.: Early Prediction of Student Dropout in Higher Education Using Machine Learning Models. In: Educational Data Mining (2024). [10]      Ennibras, F., Aoula, E., Bouihi, B.: AI in Preventing Dropout in Distance Higher Education: A Systematic Literature Review. In: 2024 4th International Conference on Innovative Research in Applied Science, Engineering and Technology (2024). doi:10.1109/IRASET60544.2024.10548954