Optimasi Model MLP Menggunakan Hyperparameter Tuning Randomsearchcv Dalam Prediksi Resiko Penyakit Jantung

Muhammad Hamdi; Djoko Rahardjo

doi:10.29408/edumatic.v9i3.32786

Authors

Muhammad Hamdi Program Studi Ilmu Komputer, Universitas Bumigora
Djoko Rahardjo Program Studi Ilmu Komputer, Universitas Bumigora

DOI:

https://doi.org/10.29408/edumatic.v9i3.32786

Keywords:

Hear Disease, MLP, SMOTE, Randomized Search, classification, machine learning.

Abstract

Heart disease is one of the leading causes of death worldwide, requiring accurate early detection methods to support prevention efforts. This study aims to develop and evaluate a heart disease prediction model using the Multilayer Perceptron (MLP) algorithm in three scenarios: a baseline model without data balancing and parameter optimization, the application of the Synthetic Minority Over-sampling Technique (SMOTE), and a combination of SMOTE with parameter optimization using Randomized Search Cross-Validation. The dataset used is heart_cleveland_upload.csv, which contains patient clinical attributes and heart disease condition labels. Pre-processing steps include handling missing data, encoding categorical variables, separating training and test data, and feature standardization. Experimental results show that the baseline MLP model achieves an accuracy of 0.85 and an F1 score of 0.830. The application of SMOTE improves the F1 score to 0.836 with a constant accuracy. The best performance is achieved by the MLP model combined with SMOTE and parameter optimization, with an accuracy of 0.90 and an F1 score of 0.884. Feature analysis shows that several clinical indicators have a dominant contribution in the prediction process.

References

Adegun, A. A., & Viriri, S. (2021). Deep learning-based system for heart disease prediction. Computers, Materials & Continua, 69(1), 19–34.

Ahn, J., Cho, S., & Kim, D. (2023). A robust feature scaling strategy for improving neural network convergence. IEEE Transactions on Neural Networks and Learning Systems, 34(9), 6589–6601.

Ali, S., Hassan, A., & Tariq, M. (2022). Enhancing the performance of neural networks using advanced normalization techniques. Expert Systems with Applications, 204, 117597. https://doi.org/10.1016/j.eswa.2022.117597

Aryuni, M., et al. (2023). Imbalanced learning in heart disease categorization: Improving minority class prediction accuracy using the SMOTE algorithm. International Journal of Fuzzy Logic and Intelligent Systems, 23(2), 140–151. https://doi.org/10.5391/IJFIS.2023.23.2.140

Deepika, D. (2022). Heart disease prediction using machine learning techniques for earlier diagnosis: Heart disease classified using novel hybrid MLP-EBMDA method. Expert Systems with Applications, 174, 102915. https://doi.org/10.1016/j.eswa.2021.100915

Firdaus, R., Mualfah, D., & Hasanah, J. S. (2023). Klasifikasi Multi-Class Penyakit Jantung dengan SMOTE dan Pearson’s Correlation menggunakan MLP. Jurnal CoSciTech (Computer Science and Information Technology), 4(1). https://doi.org/10.37859/coscitech.v4i1.4769

Gao, C., Chen, Q., & Li, Y. (2021). Handling imbalanced medical datasets using hybrid resampling methods. BMC Medical Informatics and Decision Making, 21(1), 1–12. https://doi.org/10.1186/s12911-021-01617-0

García-Floriano, A., Cruz, J., & Castejón-Limas, M. (2023). Machine learning approaches for handling imbalanced clinical data. Artificial Intelligence in Medicine, 144, 102670. https://doi.org/10.1016/j.artmed.2023.102670

Ibrahim, H., Omar, M., & Hassan, R. (2023). A machine learning framework for identifying key predictors of heart disease. Scientific Reports, 13, 2321. https://doi.org/10.1038/s41598-023-29557-1

Karthikeyan, R., & Ranganathan, K. (2021). Impact of missing data imputation techniques on heart disease prediction. Journal of King Saud University – Computer and Information Sciences, 33(6), 701–709. https://doi.org/10.1016/j.jksuci.2019.02.002

Khan, A., Rahman, S., & Al Mamun, M. (2024). Improving sensitivity in cardiovascular disease prediction models using optimized neural networks. Computers in Biology and Medicine, 170, 107832. https://doi.org/10.1016/j.compbiomed.2024.107832

Kose, U., Arslan, A., & Demir, I. (2023). Optimized deep neural networks for medical diagnosis: A comparative analysis. Biomedical Signal Processing and Control, 86, 104944. https://doi.org/10.1016/j.bspc.2023.104944

Kwakye, K., & Dadzie, E. (2021). Machine learning-based classification algorithms for the prediction of coronary heart diseases. arXiv.

Patel, J., & Nandhu, R. (2022). Efficient hyperparameter optimization using randomized search for neural networks. International Journal of Intelligent Systems and Applications, 14(2), 12–22.

Putri, S. K. N. A., Jumiatin, I., Sulistia, I., Saputra, N. A. B., & Wiranda, N. (2025). Penerapan hyperparameter tuning pada model klasifikasi untuk prediksi risiko penyakit jantung. MALCOM: Indonesian Journal of Machine Learning and Computer Science, 5(4), 1181–1189. https://doi.org/10.57152/malcom.v5i4.2138

Ramadhani, N., & Nugroho, A. (2022). SMOTE-based improvement for prediction of heart disease using machine learning. TELKOMNIKA (Telecommunication Computing Electronics and Control), 20(5), 1124–1132. https://doi.org/10.12928/telkomnika.v20i5.22067

Sharma, N., & Malviya, L. (2023). A hybrid deep neural net learning model for predicting coronary heart disease using randomized search cross-validation optimization. Decision Analytics Journal, 9, 100331. https://doi.org/10.1016/j.dajour.2023.100331

Supriyanto, H., Hariguna, T., & Barkah, A. S. (2025). MLP model optimization for heart attack risk prediction: A systematic literature review. Sinkron: Jurnal dan Penelitian Teknik Informatika, 9(3), 1811–18227. https://doi.org/10.33395/sinkron.v9i3.15027

Susanto, E. R., & Saputra, E. (2025). Implementation of deep learning with multilayer perceptron (MLP) for heart disease prediction using the SMOTE-ENN technique. Journal of Applied Informatics and Computing, 9(3), 1034–1041. https://doi.org/10.30871/jaic.v9i3.9337

Thapa, B., Khanal, S., & Sharma, P. (2022). Machine learning models with hyperparameter tuning for accurate disease prediction. Informatics in Medicine Unlocked, 32, 101020. https://doi.org/10.1016/j.imu.2022.101020

Wang, Y., Liu, L., & Sun, X. (2023). Systematic hyperparameter optimization of neural networks for healthcare prediction. IEEE Access, 11, 45012–45025.

Wijayanto, D., Marco, R., Sidauruk, A., & Sulistiyono, M. (2025). The effect of SMOTE and Optuna hyperparameter optimization on TabNet performance for heart disease classification. SISFOKOM, 14(2), 156–164. https://doi.org/10.32736/sisfokom.v14i2.2348

Zhai, X., Wang, J., & Li, Z. (2022). Managing class imbalance in healthcare datasets: A comprehensive review. Computer Methods and Programs in Biomedicine, 221, 106882. https://doi.org/10.1016/j.cmpb.2022.106882