Published on: May 2026
PRIVACY PRESERVING CLASSIFICATION USING NOISE ADDITION
Amitesh Bhaskar Aditya Girdhar Ankit Kumawat Dr. Shashidhar V
Article Status
Available Documents
Abstract
With large-scale applications of machine learning models trained on sensitive data, privacy-preserving classification is now one of the critical challenges. Traditional anonymity and masking methods are increasingly vulnerable to modern inference and linkage attacks. The key challenge is how to strike a proper balance between privacy and model accuracy, since adding too much noise may distort data utility.
This paper presents a lightweight, privacy-preserving framework that integrates Decision Tree classification with noise-based perturbation. We apply Gaussian, Laplacian, Uniform, and Exponential noise to data features in a systematic manner using the Wine dataset to investigate their effect on classification performance. Also, a bin-based uniform noise approach will be introduced to maintain the structural integrity while enhancing privacy, which limits the perturbation within bounded ranges.
Experimental results show that the baseline accuracy of 97.22% drops to 80% for Gaussian, 78% for Laplacian, and 85% for Exponential, while Uniform noise manages to retain 90%. After discretization, bin-based noise further improves the accuracy to 91.67%, almost comparable to that of the clean dataset. It is concluded from the results that bounded and controlled noise provides an efficient balance between privacy-utility that provides a lightweight alternative framework to the heavy frameworks such as differential privacy..
How to Cite this Paper
Bhaskar, A., Girdhar, A., Kumawat, A. & V, S. (2026). Privacy Preserving Classification using Noise Addition. International Journal of Creative and Open Research in Engineering and Management, <i>02</i>(05). https://doi.org/10.55041/ijcope.v2i5.403
Bhaskar, Amitesh, et al.. "Privacy Preserving Classification using Noise Addition." International Journal of Creative and Open Research in Engineering and Management, vol. 02, no. 05, 2026, pp. . doi:https://doi.org/10.55041/ijcope.v2i5.403.
Bhaskar, Amitesh,Aditya Girdhar,Ankit Kumawat, and Shashidhar V. "Privacy Preserving Classification using Noise Addition." International Journal of Creative and Open Research in Engineering and Management 02, no. 05 (2026). https://doi.org/https://doi.org/10.55041/ijcope.v2i5.403.
Search & Index
References
- Virupaksha, , Dondeti, V. Subspace based noise addition for privacy preserved data mining on high dimensional continuous data. J Ambient Intell Human Comput (2020). https://doi.org/10.1007/s12652-020-01881-8
- Hewage, H.W.A., Sinha, R. & Naeem, M.A. Privacy-preserving data (stream) mining techniques and their impact on data mining accuracy: a systematic literature review. Artif Intell Rev 56, 10427–10464 (2023). https://doi.org/10.1007/s10462-023-10425-3
- Mina Alishahi, Vahideh Moghtadaiee, Hojjat Navidan,Add noise to remove noise: Local differential privacy for feature selection,Computers & Security,Volume 123,2022,102934,ISSN 0167-4048,https://doi.org/10.1016/j.cose.2022.102934
- Sarkar, , Shinde, S., Shedge, R. (2024). Elevating Privacy in Recommendation Systems with Hybrid Noise in Local Differential Privacy. In: Tripathi, A.K., Anand, D., Nagar, A.K. (eds) Proceedings of World Conference on Artificial Intelligence: Advances and Applications. WCAIAA 2024. Algorithms for Intelligent Systems. Springer, Singapore. https://doi.org/10.1007/978-981-97-4496-1_19
- Xianlin Wu, Yuwen Chen, Haiyang Yu, Zhen Yang,Privacy-preserving federated learning based on noise addition,Expert Systems with Applications,Volume 267,2025,126228,ISSN
0957-4174,https://doi.org/10.1016/j.eswa.2024.12622 8.
- Jain, S.K., Kesswani, N. A noise-based privacy preserving model for Internet of Things. Complex Intell. 9, 3655–3679 (2023).
https://doi.org/10.1007/s40747-021-00489-5
- Zhao, Y. Chen and W. Zhang, "Differential Privacy Preservation in Deep Learning: Challenges, Opportunities and Solutions," in IEEE Access, vol. 7,
- 48901-48911, 2019, doi: 10.1109/ACCESS.2019.2909559. keywords: {Deep
learning;Training;Data models;Hidden Markov models;Differential privacy;Privacy;Deep learning;differential privacy;privacy attacks},
[8]Z. Liu, J. Guo, W. Yang, J. Fan, K. -Y. Lam and J. Zhao, "Privacy-Preserving Aggregation in Federated Learning: A Survey," in IEEE Transactions on Big Data, doi:10.1109/TBDATA.2022.3190835.
[9]Zhao, JZ., Wang, XW., Mao, KM. et al. Correlated Differential Privacy of Multiparty Data Release in Machine Learning. J. Comput. Sci. Technol. 37, 231–251(2022).https://doi.org/10.1007/s11390-021-1754-5
[10]X. Li, Y. Chen, C. Wang and C. Shen, "When Deep Learning Meets Differential Privacy: Privacy,Security, and More," in IEEE Network, vol. 35, no. 6, pp. 148-155,November/December 2021, doi: 10.1109/MNET.001.2100256.
Ethical Compliance & Review Process
- •All submissions are screened under plagiarism detection.
- •Review follows editorial policy.
- •Authors retain copyright.
- •Peer Review Type: Double-Blind Peer Review
- •Published on: May 14 2026
This article is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. You are free to share and adapt this work for non-commercial purposes with proper attribution.