University student dropout prediction improved by feature selection with Multilayer Perceptron Neural Network
Article Sidebar
Main Article Content
Abstract
The objectives of this research were to analyze the factors affected university student dropout prediction, to compare the efficiency of the prediction model, and to improve the model efficiency by using feature selection with multilayer perceptron neural network. Data used in this research were collected from the Office of Student Registration and Process, Division of Educational Service, Ubon Ratchathani University. They consisted of 3 parts: (1) 1,029 records of basic data, (2) 6,826 records of semester data, and (3) 29,790 records of grade data. After data preparation process by CRISP-DM method, 882 records remained with 14 attributes. The models of university student dropout prediction were developed using (1) Decision Tree (C4.5), (2) Naive Bayes, (3) Multilayer Perceptron Neural Network, and (4) Support Vector Machine integrated with feature selection methods including (1) Gain Ratio, (2) Chi-Square, and (3) Correlation-based Feature Selection (CFS). The efficiency of the generated models was measured by 10-Folds Cross Validation to compare (1) accuracy, (2) precision, (3) recall, and (4) mean absolute error (MAE). The experiment results revealed that the key factors to predict the university student dropout were 5 attributes including (1) grade point average (GPA), (2) GPA of courses in the student’s faculty, (3) GPA of courses outside the student’s faculty, (4) student loan or scholarship, and (5) student graduation status (target class). The best model for predicting university student dropout model was developed by Multilayer Perceptron Neural Network improved with Correlation-based Feature Selection. The accuracy of the prediction model, after model development with parameter tuning, was 90.39%. The results indicated that the feature selection could improve the efficiency of the prediction model developed by Neural Network for predicting university student dropout accurately. The generated model can further be used to develop the system for predicting university student dropout.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
The Journal of Science and Science Education (JSSE) retain the right of all articles published in JSSE. The coresponding author or the authorized person on behalf of the authors must send the complete Copyright Transfer Form to JSSE before any article get published in JSSE.
Copyright Transfer Form
The JSSE request the coresponding author or the authorized person on behalf of the authors upload the manuscript under the together with the Copyright Transfer Form under the supplementary data. The guidline for uploading both manuscript and Copyright Transfer Form is shown below:
1. Upload the manuscript in the sub-menu, Article Component > Article Text.
2. Upload the the Copyright Transfer Form in the sub-menu, Article Component > Other.
Download Copyright Transfer Form
References
Boonprasom, C. and Sanrach, C. (2018). Predictive analytic for student dropout in undergraduate using data mining technique (in Thai). Technical Education Journal King Mongkut’s University of Technology North Bangkok, 9(1), 142-151.
Habusaya, S. and Ditcharoen, N. (2020). Analysis of ICD-10 coding errors in 43 files database systems for medical record department using data mining techniques. KKU Science Journal, 48(1), 142 – 155.
Iamprik, C. and Sudadet, K. (2017). Predicting student’s data for performance by data mining. Proceedings of The Thirteenth National Conference on Computing and Information Technology NCCIT2017 (pp. 32-37).
Pacharawongsakda, E. (2014). An introduction to data mining techniques (in Thai). Bangkok: Asia Digital Press Company Limited.
Paruechanon, P. and Sriurai, W. (2018). Applying association rule to risk analysis for dropout students of Information Technology Department (in Thai). Journal of Science and Science Education, 1(2), 123-133.
Pavya, K., and Srinivasan, D. B. (2017). Feature Selection Techniques in Data Mining: A Study. International Journal of Scientific Development and Research (IJSDR), 2(6), 594-598.
Pheunpha, P. (2020). Drop-out Factors of Students of Business Administration Faculty (in Thai). Journal of Education, Mahasarakham University. 14(2), 144-158.
Prakobpol, T. (2009). Artificial Neural Networks. HCU Journal, 12(24). 73-87.
Rawengwan, P. and Seresangtakul, P. (2017). A model for forecasting educational status of students (in Thai). Proceedings of The National and International Graduate Research Conference 2017. (pp 273 – 283). Khon Kaen University.
Sittichat, S. (2017). Study of Educational Attributes Using Data Mining Technique (in Thai). Information Technology Journal, 13(2), 20–28.
Sriurai, W. (2014). Patients Classification of Metabolic Syndrome Using Feature Selection and Artificial Neural Network (in Thai). Srinakharinwirot Science Journal, 30(1), 91-102.
Techapanurak, E. (2021). Bayesian Neural Network (in Thai). Retrieved September 25, 2021, from medium.com : https://medium.com/@dopplerz/bayesian-neural-network-ตอนที่-1-ทฤฎีความน่าจะเป็นแบบเบย์ๆ-99deeab8c206
Ubon Ratchathani University. (2018). Rules and Regulations of Ubon Ratchathani University on Bachelor’s Degree Education B.E. 2561 (2018) (in Thai). Retrieved December 30, 2020, from https://www.ubu.ac.th /web/files_up/46f20 19042610283611.pdf.