Home / Current Issue / Paper 1718180
Heart Failure Classification Using Support Vector Machine with Firefly Optimization
Subject area: Science,Engineering and Technology · Area of research: Vector Machine
DOI: https://doi.org/10.64388/IREV9I11-1718180
Abstract
Heart failure remains a major global cause of death, making accurate mortality prediction crucial for effective clinical decision-making. This study proposes a mortality classification model using a Support Vector Machine (SVM) optimized with the Firefly Algorithm (FA). This approach overcomes the limitations of conventional machine learning methods that rely on manual hyperparameter tuning and frequently produce suboptimal performance when applied to complex clinical datasets. The dataset contains 299 heart failure patient records with 12 clinical attributes and one target label. Preprocessing steps include winsorizing for outlier, feature standardization, and handling class imbalance using the Synthetic Minority Oversampling Technique (SMOTE). FA is applied to automatically optimize the SVM hyperparameters, specifically cost (C) and gamma (γ), to improve classification performance. The proposed FA–SVM model achieved 90.24% test accuracy and a recall score of 0.93 for deceased patients, demonstrating strong capability in detecting high-risk cases. These results indicate that combining FA with SVM improves predictive accuracy and model stability, showing promise as a clinical decision support tool for early heart failure risk detection.
Keywords
Firefly Algorithm, Heart Failure Classification, Machine Learning, Support Vector Machine, FA-SVM
References
[1] Shams P; Malik A; Chhabra L, Heart Failure (Congestive Heart Failure). Treasure Island (FL): StatPearls Publishing, 2025. [Online]. Available: https://www.ncbi.nlm.nih.gov/books/NBK430873/
[2] Beghini et al., “2024 Update in Heart Failure,” ESC Hear. Fail., vol. 12, no. 1, pp. 8–42, 2025, doi: 10.1002/ehf2.14857.
[3] J. E. Lee and S. Y. Hwang, “Poor Nutritional Status Predicts Mortality in Patients with Acute Heart Failure Admitted to the Cardiac Intensive Care Unit: A Retrospective Cohort Study,” Asian Nurs. Res. (Korean. Soc. Nurs. Sci)., vol. 19, no. 1, pp. 53–59, 2025, doi: 10.1016/j.anr.2024.12.002.
[4] S. Nafisah, N. Nuril Inayah, and B. Yusuf, “Penyebab dan Perkembangan Penyakit Jantung Koroner,” J. Forum Kesehat. Media Publ. Kesehat. Ilm., vol. 14, no. 1, 2024.
[5] Dimsyiar M Al Hafiz, Khoirul Amaly, Javen Jonathan, M Teranggono Rachmatullah, and Rosidi, “Sistem Prediksi Penyakit Jantung Menggunakan Metode Naive Bayes,” J. Rekayasa Elektro Sriwij., vol. 2, no. 2, pp. 151–157, 2021, doi: 10.36706/jres.v2i2.29.
[6] A. Muhammad, R. Amril, and M. Siregar, “Penerapan Algoritma K-Nearest Neighbor Untuk Prediksi Kematian Akibat Penyakit Gagal Jantung,” vol. III, no. 1, 2022, [Online]. Available: https://www.kaggle.com/andrewmvd/heart-failure-clinical-data.
[7] L. N. Farida and S. Bahri, “Klasifikasi Gagal Jantung menggunakan Metode SVM (Support Vector Machine),” Komputika J. Sist. Komput., vol. 13, no. 2, pp. 149–156, 2024, doi: 10.34010/komputika.v13i2.11330.
[8] Riyanto, H. Destiana, T. Prihatin, and G. Wijaya, “Mengoptimalkan Prediksi Gagal Jantung Dengan Kombinasi,” vol. 8, no. 1, pp. 103–111, 2025.
[9] L. Budianti and Suliadi, “Metode Weighted Random Forest dalam Klasifikasi Prediksi Kelangsungan Hidup Pasien Gagal Jantung,” Bandung Conf. Ser. Stat., vol. 2, no. 2, pp. 103–110, 2022, doi: 10.29313/bcss.v2i2.3318.
[10] W. Lestari and S. Sumarlinda, “Studi Komparatif Model Klasifikasi Kerentanan Penyakit Jantung Menggunakan Algoritma Machine Learning,” SATIN - Sains dan Teknol. Inf., vol. 9, no. 1, pp. 107–115, 2023, doi: 10.33372/stn.v9i1.918.
[11] Chicco and G. Jurman, “Machine learning can predict survival of patients with heart failure from serum creatinine and ejection fraction alone,” BMC Med. Inform. Decis. Mak., vol. 20, no. 1, pp. 1–16, 2020, doi: 10.1186/s12911-020-1023-5.
[12] Senthil, S., et al. “A Comparative Analysis on the Prediction of Heart Failure Using Machine Learning Algorithms.” Proceedings of the International Conference on Machine Learning, Communication, Systems and Intelligence (ICMCSI), 2024, pp. 206–211.
[13] M. K. Biddinika, A. Masitha, V. Arfiana, and N. Fatimah, “Machine Learning Techniques for Heart Disease Prediction Using a Multi-Algorithm Approach,” vol. 12, no. 2, pp. 149–158, 2024.
[14] Husni, A. Amin, S. Wibisono, and E. Lestariningsih, “Optimizing Heart Disease Prediction with Random Forest and Ensemble Methods,” vol. 11, no. 1, pp. 80–90, 2025.
[15] R. Guido, S. Ferrisi, and D. Lofaro, “An Overview on the Advancements of Support Vector Machine Models in Healthcare Applications : A Review,” 2024.
[16] N. H. Ovirianti, M. Zarlis, and H. Mawengkang, “Support Vector Machine Using A Classification Algorithm,” vol. 6, no. 3, pp. 2103–2107, 2022.
[17] R. Sitepu, A. A. Rohmawati, and S. S. Prasetyowati, “The Analysis of Support Vector Machine (SVM) on Monthly Covid-19 Case Classification,” Intl. Journal on ICT, vol. 8, no. 2, pp. 40–52, Dec. 2022, doi: 10.21108/ijoict.v8i2.671.
[18] X. Wang, H. Liu, A. G. Hussien, G. Hu, and L. Zhang, “Enhanced Particle Swarm Optimization Algorithm Based on SVM Classifier for Feature Selection,” 2025, doi: 10.32604/cmes.2025.058473.
[19] Famuyiwa, Kolawle Samuel A. , et al. “Development of Palmvein Recognition System Using Fire Fly Algorithm.” Iconic Research and Engineering Journals, vol. 6, no. 1, 2022, pp. 356–365.
[20] Y. Yao, et al., “Improved Firefly Algorithm -Optimized ResNet18 for Partial Discharge Pattern Recognition Within Small-Sample Scenarios,” pp. 1–23, 2025.
[21] Wirth, Rüdiger and Jochen Hipp. “CRISP-DM: Towards a Standard Process Model for Data Mining,” 2000.
[22] Resmi, V., Vijayalakshmi Soundararajan, and R. Chandrabose. “An Effective Software Project Effort Estimation System Using Optimal Firefly Algorithm.” Cluster Computing, vol. 22, 2019.
How to cite this paper
@article{1718180,
author = {Khoirunnisa, Abiyyu Rasyiq Muhadzzib, Yazid Hilmi Allamsyah, Detty Purnamasari, Ulfa Hidayati},
title = {Heart Failure Classification Using Support Vector Machine with Firefly Optimization},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {11},
pages = {3907-3913},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1718180.pdf},
abstract = {Heart failure remains a major global cause of death, making accurate mortality prediction crucial for effective clinical decision-making. This study proposes a mortality classification model using a Support Vector Machine (SVM) optimized with the Firefly Algorithm (FA). This approach overcomes the limitations of conventional machine learning methods that rely on manual hyperparameter tuning and frequently produce suboptimal performance when applied to complex clinical datasets. The dataset contains 299 heart failure patient records with 12 clinical attributes and one target label. Preprocessing steps include winsorizing for outlier, feature standardization, and handling class imbalance using the Synthetic Minority Oversampling Technique (SMOTE). FA is applied to automatically optimize the SVM hyperparameters, specifically cost (C) and gamma (γ), to improve classification performance. The proposed FA–SVM model achieved 90.24% test accuracy and a recall score of 0.93 for deceased patients, demonstrating strong capability in detecting high-risk cases. These results indicate that combining FA with SVM improves predictive accuracy and model stability, showing promise as a clinical decision support tool for early heart failure risk detection.},
keywords = {Firefly Algorithm, Heart Failure Classification, Machine Learning, Support Vector Machine, FA-SVM},
month = {May},
doi = {https://doi.org/10.64388/IREV9I11-1718180}
}