Differentially Private Ensemble Learning Based Logistic Regression Model for Heart Disease Prediction
DOI:
https://doi.org/10.69996/jsihs.2025008Keywords:
Differential Privacy, Logistic Regression, Heart Disease Prediction, Data Privacy, Machine LearningAbstract
Recently, there has been a lot of interest in Differential Privacy as a generic framework forprotecting personal data that is used and released for different purposes. Moreover, it has been suggestedas a suitable paradigm for protecting medical data. The proposed work has two parts. The first, partdescribes a differentially private Logistic regression model for heart disease prediction. The proposedmodel utilizes the noise-adding mechanism of the Differential Privacy technique to protect the user'shealthcare data. The model provides the training data privacy, model parameters privacy, and modeloutput privacy. The noises such as Laplacian, Gaussian, and Exponential noise with different scales areadded to the data and analyzed for optimum data privacy and utility. The differentially private model is aweak models and its classification accuracy is less. So, we added an ensemble learning concept in thesecond part of our work in order to improve the performance of the proposed differentially private logisticregression model. We have used Differentially Private Random Forest and Differential Private LinearSupport Vector Machine (SVM) models along with differential private logistic model. The Majorityvoting is used to find the final prediction output in the ensemble learning. The proposed model'sperformance is measured using various parameters such as prediction accuracy, and loss. The Laplacenoise-based mechanism provides better performance than the other two noise-based Differential Privacy.The proposed model performance in protecting the training data, model parameters, and output dataprivacy is analyzed and proved that the Laplace noise-based Differentially Private Logistic Regressionmodel is one of the simple choices to protect sensitive medical data processing. We also analyzed theprediction accuracy of differentially private SVM, and differentially private Random forest (Laplacenoise-based private models) with an ensemble learning-based logistic regression model and proved thatDifferentially private ensemble learning Logistic regression model produces an accuracy of 86.21%which is nearly equal to the model accuracy without privacy.
References
[1] C. Dwork, F. McSherry, K. Nissim, and A. Smith, “Calibrating noise to sensitivity in private data analysis,” Theory of Cryptography, pp. 265–284, 2006.
[2] M. Abadi, A. Chu, I. Goodfellow, H.B. McMahan, I. Mironov et al., “Deep learning with differential privacy,” In Proceedings of the 2016 ACM SIGSAC conference on computer and communications security, pp. 308-318, 2016.
[3] F.K. Dankar and K. El Emam, “Practicing differential privacy in health care: A review,” Trans. Data Priv., vol.6, no.1, pp.35-67, 2013.
[4] A. Dyda, M. Purcell, S. Curtis, E. Field, P. Pillai et al., “Differential privacy for public health data: An innovative tool to optimize information sharing while protecting data confidentiality,” Patterns, vol.2, no.12, 2021.
[5] H. Nori, R. Caruana, Z. Bu, J.H. Shen and J.Kulkarni, “Accuracy, interpretability, and differential privacy via explainable boosting,” In International conference on machine learning, pp. 8227-8237, 2021.
[6] J. Dong, A. Roth and W.J.Su, “Gaussian differential privacy,” arXiv preprint arXiv:1905.02383, 2019.
[7] R. Danger, “Differential Privacy: What is all the noise about?,” arXiv preprint arXiv:2205.09453, 2022.
[8] M. Abadi, A. Chu, I. Goodfellow, H.B. McMahan, I. Mironov et al., “Deep learning with differential privacy,” In Proceedings of the 2016 ACM SIGSAC conference on computer and communications security, pp. 308-318, 2016.
[9] K. Mivule, C. Turner and S.Y.Ji, “Towards a differential privacy and utility preserving machine learning classifier,” Procedia Computer Science, vol.12, pp.176-181, 2012.
[10] N. Papernot, M. Abadi, U. Erlingsson, I. Goodfellow and K.Talwar, “Semi-supervised knowledge transfer for deep learning from private training data,” arXiv preprint arXiv:1610.05755, 2016.
[11] N. Ponomareva, H. Hazimeh, A. Kurakin, Z. Xu, C. Denison et al., “How to dp-fy ml: A practical guide to machine learning with differential privacy,” Journal of Artificial Intelligence Research, vol.77, pp.1113-1201, 2023.
[12] S. Ioffe and C. Szegedy, “Batch normalization: Accelerating deep network training by reducing internal covariate shift,” In International conference on machine learning, pp. 448-456, 2015.
[13] M. Rigaki and S.Garcia, “A Survey of Privacy Attacks in Machine Learning,” arXiv 2020. arXiv preprint arXiv:2007.07646, 2020.
[14] X. Zhang, F. Yang, Y. Guo, H. Yu, Z. Wang and Q.Zhang, “Adaptive differential privacy mechanism based on entropy theory for preserving deep neural networks,” Mathematics, vol.11, no.2, pp.330, 2023.
[15] M. Abadi, A. Chu, I. Goodfellow, H.B. McMahan, I. Mironov et al., “Deep learning with differential privacy,” In Proceedings of the 2016 ACM SIGSAC conference on computer and communications security, pp. 308-318, 2023.
[16] N. Phan, X. Wu, H. Hu and D. Dou, “Adaptive laplace mechanism: Differential privacy preservation in deep learning,” In 2017 IEEE international conference (on data mining (ICDM), pp. 385-394, 2017.
[17] T. Ha, T.K. Dang, H. Le and T.A.Truong, “Security and privacy issues in deep learning: a brief review,” SN Computer Science, vol.1, no.5, pp.253, 2020.
[18] Y. Wang, Q.Wang, L. Zhao and C. Wang, “Differential privacy in deep learning: Privacy and beyond,” Future Generation Computer Systems, 2023.
[19] J. Zhao, Y. Chen and W. Zhang, “Differential privacy preservation in deep learning: Challenges, opportunities and solutions,” IEEE Access, vol.7, pp.48901-48911, 2019.
[20] P.Jain, M. Gyanchandani and N. Khare, “Differential privacy: its technological prescriptive using big data,” Journal of Big Data, vol.5, no.1, pp.1-24, 2018.
[21] F.K.Dankar and K.Emam, “Practicing differential privacy in health care: A review,” Transactions on Data Privacy, vol.6, no.1, pp.35–67, 2013.
[22] H. Cho, S. Simmons, R. Kim and B. Berger, “Privacy-preserving biomedical database queries with optimal privacy-utility trade-offs,” Cell Systems, vol.10, pp.408–416, 2020.
[23] J. Jordon, J. Yoon and M. van der Schaar, “PATE-GAN: Generating synthetic data with differential privacy guarantees,” In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6–9, 2019.
[24] B.K. Beaulieu-Jones, Z.S. Wu, C. Williams, R. Lee, S.P. Bhavnani et al., “Privacy-preserving generative deep neural networks support clinical data sharing,” Circulation: Cardiovascular Quality and Outcomes, vol.12, pp.5122, 2019.
[25] B. Ghazi, N. Golowich, R. Kumar, P. Manurangsi and C. Zhang, “Deep learning with label differential privacy,” In Advances in neural information processing systems, 2021.
[26] V.S. Naresh and M. Thamarai, “Privacy‐preserving data mining and machine learning in healthcare: Applications, challenges, and solutions,” Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, vol.13, no.2, pp.e1490, 2023.
[27] T. Bhaskar, M.N. Narsaiah and M. Ravikanth, “Central Medical Centre Healthcare Data Security with Lightweight Blockchain Model in IoT Sensor Environment,” Journal of Sensors, IoT and Health Sciences, vol.1, no.01, pp.15-26, 2023.
[28] S. Venkatramulu, K.V. Kumar, M.S. Waseem, S. Mahveen, V. Vaidya et al., “A secure blockchain based student certificate generation and sharing system,” Journal of Sensors, IoT and Health Sciences, vol.2, no.01, pp.17-27, 2024.
[29] I. Ilham, “Enhancing Cardiovascular Disease Prediction Accuracy through an Ensemble Machine Learning Approach,” International Journal of Artificial Intelligence in Medical Issues, vol.2, no.2, pp.95-103, 2024.
[30] R. Mahaveerakannan, “Ensemble Deep Learning for Enhanced Human Microbiome-Based Disease Prediction: A Comprehensive Analysis on Public Datasets,” In 2024 8th International Conference on Computational System and Information Technology for Sustainable Solutions (CSITSS), pp. 1-4,2024
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Journal of Sensors, IoT & Health Sciences (JSIHS,ISSN: 2584-2560)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Fringe Global Scientific Press publishes all the papers under a Creative Commons Attribution-Non-Commercial 4.0 International (CC BY-NC 4.0) (https://creativecommons.org/licenses/by-nc/4.0/) license. Authors have the liberty to replicate and distribute their work. Authors have the ability to use either the whole or a portion of their piece in compilations or other publications that include their own work. Please see the licensing terms for more information on reusing the work.
