A Study on Privacy Data Protection in AI Training

Xinru Yang

doi:10.47297/taposatWSP2633-456924.20260704

PDF
导出
分享
收藏
专辑

A Study on Privacy Data Protection in AI Training

7卷第4期

浏览量: 0 Downloads: 0

A Study on Privacy Data Protection in AI Training
A Study on Privacy Data Protection in AI Training
School of Information and Intelligent Engineering, Zhejiang Wanli College, Ningbo, Zhejiang, 315000, China
2026年7卷第4期页码：129-134
纸质出版：2026-04-30
DOI：10.47297/taposatWSP2633-456924.20260704
Accepted：

Scan QR Code

Xinru Yang. (2026). A Study on Privacy Data Protection in AI Training. 科技理论与实践（英文版）, 7(4), 129-134.

Yang Xinru.A Study on Privacy Data Protection in AI Training[J].Theory and Practice of Science and Technology,2026,07(04):129-134.
Xinru Yang. (2026). A Study on Privacy Data Protection in AI Training. 科技理论与实践（英文版）, 7(4), 129-134. DOI： 10.47297/taposatWSP2633-456924.20260704.

Yang Xinru.A Study on Privacy Data Protection in AI Training[J].Theory and Practice of Science and Technology,2026,07(04):129-134. DOI： 10.47297/taposatWSP2633-456924.20260704.

摘要

Abstract

AI technology deeply penetrates into key fields such as healthcare and finance

and its model training relies on a large amount of real data containing sensitive information

and the risk of privacy leakage is highlighted in data sharing and collaborative training

and there are obvious loopholes in the traditional protection techniques

and emerging technologies such as differential privacy and homomorphic encryption have become the core paths to solve the problem

but they are faced with the problems of technological adaption and efficiency loss. This study takes the privacy data protection in AI training as the theme

adopts literature research

comparative research and inductive analysis method

systematically combs through the application status quo and core pain points of emerging privacy protection technologies

clarifies the research progress at home and abroad

constructs a theoretical framework of privacy protection adapted to AI training

provides theoretical support and practical references for the security protection of data in sensitive areas

and proposes the direction of technological innovation and development in combination with industrial needs. It also puts forward the direction of technological innovation and development

which makes up for the shortcomings of the existing research in terms of comprehensiveness and scene adaptability.

关键词

Keywords

references

Wu Zijian. Research and application of data privacy security protection based on differential privacy [D]. Liaoning University of Science and Technology,2022.

Gao Yuan. Research on secure data storage in cloud computing network based on homomorphic encryption[J]. Network Security Technology and Application,2025,(11):37-39.

Owusu-Agyemeng K ,Qin Z ,Xiong H , et al.MSDP: multi-scheme privacy-preserving deep learning via differential privacy[J].Personal and Ubiquitous Computing,2021,27(2):1-13.

Khanh T D ,T. P T T .A Pragmatic Privacy-Preserving Deep Learning Framework Satisfying Differential Privacy[J].SN Computer Science,2023,5(1).

Olabim M, Greenfield A, Barlow A. A differential privacy-based approach for mitigating data theft in ransomware attacks[J]. Authorea Preprints, 2024.

Victor K ,Alla L ,Alexander V . Homomorphic Encryption within Lattice-Based Encryption System[J].Procedia Computer Science,2021,186309-315.

Schwartzentruber J .Security, privacy, and generative AI[J].InfoWorld.com,2023.

Anitha K K ,Avinash S ,Chinmay C , et al.Preserving Health Care Data Security and Privacy Using Carmichael's Theorem-Based Homomorphic Encryption and Modified Enhanced Homomorphic Encryption Schemes in Edge Computing Systems.[J].Big data,2021,10(1).

David E ,Eldon S .AI Technologies, Privacy, and Security [J].Frontiers in Artificial Intelligence,2022,5826737-826737.

Yan Zhao. Research on data differential privacy based on smart medical applications[D]. Nanjing University of Posts and Telecommunications,2023.

Wu, N.B Research on differential privacy equilibrium optimization model and algorithm[D]. Guizhou University,2021.

Liu J. Research on Differential Privacy Protection Methods for Integrated Learning[D]. Guangxi Normal University,2021.

Lv Kangyu. Research on privacy protection of logistic regression training based on homomorphic encryption[D]. Dalian University of Technology,2020.

J. Jiang. Research on searchable encryption algorithm based on Paillier homomorphic encryption[D]. Hunan University,2020.

Fang Mingyang. Research on federated learning privacy encryption technique based on homomorphic encryption[D]. East China Normal University,2023.

Jia H. X. Development of a medical data model training platform based on searchable encryption[D]. Jinan University,2022.

Wu, Danhua. Research on deep learning model image data training method based on differential privacy[D]. Guilin University of Technology,2023.

Li Y,He Chunlin. Data-differential privacy-preserving stochastic gradient descent algorithm for deep neural network training[J]. Computer Application and Software,2020,37(04):252-259.

Juanping Zhang,Hailiang Zhang. Research on issues related to big data and data privacy protection[J]. Software,2023,44(02):81-84.

Jestine P ,Selva M S M A ,William M , et al.Privacy-Preserving Collective Learning With Homomorphic Encryption[J].IEEE ACCESS,2021,9132084-132096.

Yarong L .Data privacy protection based on homomorphic encryption[J].Journal of Physics: Conference Series,2021,2037(1).

文章被引用时，请邮件提醒。

Submit

A Comparative Review of Privacy Protection Techniques in Federated Learning

A Review of the Principles and Typical Applications of Differential Privacy Technology

A Survey of Privacy-Preserving Techniques in Federated Graph Neural Networks

Privacy Protection of Clustering High-Dimensional Data Based on African Vultures

A Review of Human-Machine Interaction Methods for Consumer Drones in the Context of the Low-Altitude Economy

Related Author

Jiayi Li School of Data Science and Artificial Intelligence, Wenzhou Institute of Technology

Ketong Liu Reading College, Nanjing University of Information Science and Technology

Lingyu Wang School of Arts and Sciences, China University of Petroleum-Beijing at Karamay

Chenglin Zhang School of Information & Electronics Technology, Jiamusi University

Lili He School of Information & Electronics Technology, Jiamusi University

Related Institution

School of Data Science and Artificial Intelligence, Wenzhou Institute of Technology

Reading College, Nanjing University of Information Science and Technology

School of Arts and Sciences, China University of Petroleum-Beijing at Karamay

School of Information & Electronics Technology, Jiamusi University

⁰