Prediction of New Student Registration Trends at Religious Higher Education Institutions Using Machine Learning and Deep Learning Approaches with SHAP Interpretation

Alit Fajar Kurniawan; Umi Laili Yuhana; Bilqis Amaliah

doi:10.59261/bustechno.v7i4.770

Authors

Alit Fajar Kurniawan Institut Teknologi Sepuluh Nopember
Umi Laili Yuhana Institut Teknologi Sepuluh Nopember
Bilqis Amaliah Institut Teknologi Sepuluh Nopember

DOI:

https://doi.org/10.59261/bustechno.v7i4.770

Keywords:

CNN-LSTM, Feature Engineering, Multivariate Time Series Forecasting, New Student Registration, SHAP

Abstract

Background: Reliable forecasts of new student registration are essential for religious higher education institutions because admission volumes affect capacity planning, resource allocation, promotional strategies, and program sustainability. Historical admission data are often used descriptively rather than to generate forward-looking and interpretable estimates.

Objective: This study developed an explainable multivariate time-series framework for forecasting new student registration by integrating domain-based feature engineering, comparative machine-learning and deep-learning models, CNN-LSTM forecasting, and SHAP interpretation.

Methods: Historical admission records from 40 study programs at UIN Ar-Raniry for 2017–2025 were analyzed, comprising 360 program-year observations. The workflow included data validation and preprocessing, construction of share, pressure, and lag features, two-year sequence construction, temporal train-validation-test splitting, train-only scaling, comparative model evaluation, 2026 forecasting, and SHAP-based global and local interpretation.

Results: CNN-LSTM achieved the best average performance across five iterations, with an MAE of 6.0394, an RMSE of 8.6185, a MAPE of 27.7548%, and an R² of 0.9167. Its advantage is consistent with the structure of the data: convolutional layers extract local relationships among multivariate admission features, while LSTM captures temporal dependencies. SHAP showed that recent historical registration variables, especially reg_span_lag1, reg_um_lag1, and total_reg_lag1, together with capacity-related features such as jml_dt, were major contributors to the forecasts.

Conclusion: The study demonstrates that combining domain-informed feature engineering, CNN-LSTM, and SHAP provides both accurate and interpretable registration forecasts. The framework contributes an explainable decision-support approach for admission planning, quota setting, resource allocation, and targeted recruitment in religious higher education.

References

Agustina, I., Mulyani, Y., Septiana, T., & Mardiana, M. (2022). Analisis pengembangan model prediksi kesuksesan Kickstarter menggunakan Algoritma Backpropagation dan Random Forest. Jurnal Informatika Dan Teknik Elektro Terapan, 10(3), 173–182. https://doi.org/10.23960/Jitet.V10i3.2742

Bank, W. (2026). School Enrollment, Tertiary (% Gross) (Se.Ter.Enrr), World Development Indicators. World Bank. https://data.worldbank.Org/Indicator/Se.Ter.Enrr

Bousnguar, H., Najdi, L., & Battou, A. (2022). Forecasting approaches in A higher education setting. Education And Information Technologies, 27(2), 1993–2011. https://doi.org/10.1007/S10639-021-10684-Z

Chen, J., Guan, A., Du, J., & Ayush, A. (2025). Multivariate time series prediction with multi-feature analysis. Expert Systems With Applications, 268, 126302. https://doi.org/10.1016/J.Eswa.2024.126302

Farisi, I., Shadiq, J., Priyadi, W., Maulana, D., Acep, A., & Gusril, S. F. (2024). Penerapan model Recurrent Neural Network (RNN) untuk prediksi curah hujan berbasis data historis. Information System For Educators And Professionals: Journal Of Information System, 9(2), 217–226. https://doi.org/10.51211/Isbi.V9i2.3280

Harahap, D., & Alfadri, F. (2021). Ekonomi Mikro Islam (1st Ed.). Merdeka Kreasi Group.

Huang, D., Liu, Z., & Wu, D. (2024). Research on ensemble learning-based feature selection method for time-series prediction. Applied Sciences, 14(1), 40. https://doi.org/10.3390/App14010040

James, F. E., & Weese, J. L. (2022). Neural network-based forecasting of student enrollment with exponential smoothing baseline and performance analysis. In 2022 Asee Annual Conference & Exposition. Asee Conferences. https://doi.org/10.18260/1-2--41751

Ji, Y., Lei, X., Zhang, L., Heng, J., & Fan, J. (2026). Forecasting the price of gold with integrated media sentiment—a prediction framework based on online news sentiment mining with Cnn-Qrlstm. Entropy, 28(3), 271.

Kabir, M. R., Bhadra, D., Ridoy, M., & Milanova, M. (2025). LSTM–Transformer-Based Robust Hybrid Deep Learning Model For Financial Time Series Forecasting. Sci, 7(1), 7. https://doi.org/10.3390/Sci7010007

Kong, X., Chen, Z., Liu, W., Ning, K., Zhang, L., Marier, S. M., Liu, Y., Chen, Y., & Xia, F. (2025). Deep Learning For Time Series Forecasting: A Survey. International Journal Of Machine Learning And Cybernetics, 16(7–8), 5079–5112. https://doi.org/10.1007/S13042-025-02560-W

Lima, F. T., & Souza, V. M. A. (2023). A Large Comparison Of Normalization Methods On Time Series. Big Data Research, 34, 100407. https://doi.org/10.1016/J.Bdr.2023.100407

Liu, T., Schenk, C., Braun, S., & Frey, A. (2025). A Machine-Learning-Based Approach To Informing Student Admission Decisions. Behavioral Sciences, 15(3), 330. https://doi.org/10.3390/Bs15030330

Livieris, I. E., Pintelas, E., & Pintelas, P. (2020). A Cnn–Lstm Model For Gold Price Time-Series Forecasting. Neural Computing And Applications, 32(23), 17351–17360. https://doi.org/10.1007/S00521-020-04867-X

Neubauer, A., Brandt, S., & Kriegel, M. (2025). Explainable Multi-Step Heating Load Forecasting: Using Shap Values And Temporal Attention Mechanisms For Enhanced Interpretability. Energy And Ai, 20, 100480. https://doi.org/10.1016/J.Egyai.2025.100480

OECD. (2025). Education At A Glance 2025: Oecd Indicators. Oecd Publishing. https://doi.org/10.1787/1c0d9c79-En

Parisineni, S. R. A., & Pal, M. (2024). Enhancing Trust And Interpretability Of Complex Machine Learning Models Using Local Interpretable Model Agnostic Shap Explanations. International Journal Of Data Science And Analytics, 18(4), 457–466. https://doi.org/10.1007/S41060-023-00458-W

Pranolo, A., Setyaputri, F. U., Paramarta, A. K. I., Triono, A. P. P., Fadhilla, A. F., Akbari, A. K. G., Utama, A. B. P., Wibawa, A. P., & Uriu, W. (2024). Enhanced Multivariate Time Series Analysis Using Lstm: A Comparative Study Of Min-Max And Z-Score Normalization Techniques. Ilkom Jurnal Ilmiah, 16(2), 210–220. https://doi.org/10.33096/Ilkom.V16i2.2333.210-220

Puspanagara, A. L. (2025). Penerapan Explainable Ai Untuk Prediksi Performa Akademik Mahasiswa Menggunakan Random Forest Dan Shap. Infoman’s: Jurnal Ilmu-Ilmu Informatika Dan Manajemen, 19(1), 1–7.

Saputra, R. A., & Pratama, A. (2025). Implementasi Decision Tree Untuk Prediksi Harga Rumah Di Daerah Tebet. Journal Of Information System Management (Joism), 6(2), 164–170. https://doi.org/10.24076/Joism.2025v6i2.1928

Saravana, M. K., Roopa, M. S., Arunalatha, J. S., & Venugopal, K. R. (2026). Transformers For Multivariate Time Series Forecasting: Comprehensive Analysis, Challenges, Research Opportunities And Future Prospects. IEEE Access.

Shih, S. Y., Sun, F. K., & Lee, H. Y. (2019). Temporal Pattern Attention For Multivariate Time Series Forecasting. Machine Learning, 108(8–9), 1421–1441. https://doi.org/10.1007/S10994-019-05815-0

Unesco. (2026). Higher Education. Unesco. https://www.unesco.org/En/Higher-Education

Van Zyl, C., Ye, X., & Naidoo, R. (2024). Harnessing Explainable Artificial Intelligence For Feature Selection In Time Series Energy Forecasting: A Comparative Analysis Of Grad-Cam And Shap. Applied Energy, 353(Part A), 122079. https://doi.org/10.1016/J.Apenergy.2023.122079

Wahyudin, M., Husaini, M., & Sa’adah, U. N. (2025). Ekonomi Mikro Islam.

Wahyuni, A. T. (2025). Manajemen Humas Iain Curup Dalam Mempromosikan Penerimaan Calon Mahasiswa Baru Di Tahun 2024 [Institut Agama Islam Negeri (IAIN) Curup]. Https://E-Theses.Iaincurup.Ac.Id/9171/1/Anisa Tri Wahyuni Fulltext.Pdf

Wen, X., & Li, W. (2023). Time Series Prediction Based On Lstm-Attention-Lstm Model. Ieee Access, 11, 48322–48331. https://doi.org/10.1109/Access.2023.3276628.