DOI QR코드

DOI QR Code

개인정보 보호를 고려한 딥러닝 데이터 자동 생성 방안 연구

A Study of Automatic Deep Learning Data Generation by Considering Private Information Protection

  • 장성봉 (금오공과대학교 산학협력단)
  • Sung-Bong Jang (Dept. of Industry-Academy, Kumoh Institute of Technnology)
  • 투고 : 2023.10.15
  • 심사 : 2023.11.10
  • 발행 : 2024.01.31

초록

수집된 대량의 데이터셋이 딥러닝 학습데이터로 사용되기 위해서는 주민번호, 질병 정보등과 같이 민감한 개인정보는 해커에게 노출되지 않도록 값을 변경하거나 암호화해야 하고 구축된 딥러닝 모델의 구조와 일치 하도록 데이터를 재구성 해주어야 한다. 현재, 이러한 작업은 전문가에 의해 수동으로 이루어지기 때문에, 시간과 비용이 많이 소요 된다. 이러한 문제점을 해결하기 위해, 본 논문에서는 딥러닝 과정에서 개인정보 보호를 위한 데이터 처리 작업을 자동으로 수행할 수 있는 기법을 제안한다. 제안된 기법에서는 데이터 일반화에 기반한 개인정보 보호 작업을 수행하고 원형큐를 사용하여 데이터 재구성 작업을 수행한다. 제안된 기법의 타당성을 검증하기 위해, C언어를 사용하여 직접 구현하였다. 검증 결과, 데이터 일반화가 정상적으로 수행되고 딥러닝 모델에 맞는 데이터 재구성이 제대로 수행됨을 확인 할 수 있었다.

In order for the large amount of collected data sets to be used as deep learning training data, sensitive personal information such as resident registration number and disease information must be changed or encrypted to prevent it from being exposed to hackers, and the data must be reconstructed to match the structure of the built deep learning model. Currently, these tasks are performed manually by experts, which takes a lot of time and money. To solve these problems, this paper proposes a technique that can automatically perform data processing tasks to protect personal information during the deep learning process. In the proposed technique, privacy protection tasks are performed based on data generalization and data reconstruction tasks are performed using circular queues. To verify the validity of the proposed technique, it was directly implemented using C language. As a result of the verification, it was confirmed that data generalization was performed normally and data reconstruction suitable for the deep learning model was performed properly.

키워드

과제정보

이 연구는 금오공과대학교 대학 학술연구비로 지원되었음(2022년)

참고문헌

  1. W. Wei and L. Liu, "Gradient Leakage Attack Resilient Deep Learning," IEEE Transactions on Information Forensics and Security, Vol. 17, pp. 303-316,2022(DOI:10.1109/TIFS.2021.3139777)
  2. N. Bugshan, I. Khalil, M. S. Rahman, M. Atiquzzaman;X. Yi, S. Badsha, "Toward Trustworthy and Privacy-Preserving Federated Deep Learning Service Framework for Industrial Internet of Things," IEEE Transactions on Industrial Informatics, Vol. 19, No.2 pp. 1535-1547,2023(DOI: 10.1109/TII.2022.3209200)
  3. D. Mistry, M. F. Mridha, Me. Safran, S. Alfarhood, A. K. Saha, D. Che, "Privacy-Preserving On-Screen Activity Tracking and Classification in E-Learning Using Federated Learning," IEEE Access, Vol. 11, pp. 79315-79329, 2023(DOI: 10.1109/ACCESS.2023.3299331)
  4. I. Fontana, M. Langheinrich, M. Gjoreski, "GANs for Privacy-Aware Mobility Modeling," IEEE Access, Vol. 11, pp. 29250-29262, 2023(DOI:10.1109/ACCESS.2023.3260981)
  5. R. Podschwadt, D. Takabi, P. Hu, M. H. Rafiei, Z. Cai, "A Survey of Deep Learning Architectures for Privacy-Preserving Machine Learning With Fully Homomorphic Encryption," IEEE Access, Vol. 10, pp.117477-117500,2022(DOI:10.1109/ACCESS.2022.3219049)
  6. T. Zhang, T. Zhu, K. Gao, W. Zhou, P. S. Yu, "Balancing Learning Model Privacy, Fairness, and Accuracy With Early Stopping Criteria," IEEE Transactions on Neural Networks and Learning Systems, Vol. 34, No.9, pp.5557-5569,2023(DOI: 10.1109/TNNLS.2021.3129592)
  7. V. Stephanie, I. Khalil, M. S. Rahman, M. Atiquzzaman, "Privacy-Preserving Ensemble Infused Enhanced Deep Neural Network Framework for Edge Cloud Convergence," IEEE Internet of Things Journal, Vol. 10, No.5, pp.3763-3773,2023(DOI: 10.1109/JIOT.2022.3151982)
  8. M. Shateri, F. Messina, P. Piantanida, F. Labeau, "Privacy-Cost Management in Smart Meters With Mutual-Information-BasedReinforcement Learning," IEEE Internet of Things Journal, Vol. 9, No.22,pp.22389-22398,2022(DOI:10.1109/JIOT.2021.3128488)
  9. T. Zhu, D. Ye, W. Wang, W. Zhou, P. S. Yu, "More Than Privacy: Applying Differential Privacy in Key Areas of Artificial Intelligence," IEEE Transactions on Knowledge and Data Engineering, Vol. 34, No. 6,pp.2824-2843,2022(DOI:10.1109/TKDE.2020.3014246)
  10. Z. Wu, H. Wang, Z. Wang, H. Jin, Z. Wang, "Privacy-Preserving Deep Action Recognition: An Adversarial Learning Framework and A New Dataset," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 44, No. 4, pp. 2126-2139,2022(DOI: 10.1109/TPAMI.2020.3026709)
  11. L. Xiang, W. Li, J. Yang, X. Wang, B. Li, "Differentially-Private Deep Learning With Directional Noise," IEEE Transactions on Mobile Computing, Vol. 22, No. 5, pp. 2599-2612,2023(DOI:10.1109/TMC.2021.3130060)
  12. A. E. Ouadrhiri, A. Abdelhadi, "Differential Privacy for Deep and Federated Learning: A Survey," IEEE Access, Vol. 10, pp. 22359-22380,2022(DOI:10.1109/ACCESS.2022.3151670)
  13. R. Parekh, N. Patel, R. Gupta, N. K. Jadav, S. Tanwar, A. Alharbi, A. Tolba, B.-C. Neagu, M. S. Raboaca, "GeFL: Gradient Encryption-Aided Privacy Preserved Federated Learning for Autonomous Vehicles," IEEE Access, Vol. 11, pp. 1825-1839,2023(DOI:10.1109/ACCESS.2023.3233983)
  14. W. Zhang, B. Jiang, M. Li, X. Lin, "Privacy-Preserving Aggregate Mobility Data Release: An Information-Theoretic Deep Reinforcement Learning Approach," IEEE Transactions on Information Forensics and Security, Vol. 17, pp. 849-864, 2022(DOI:10.1109/TIFS.2022.3152361)
  15. X. Ma, Q. Jiang, X. Liu, Q. Pei, J. Ma, W. Lou "Learning in Your "Pocket": Secure Collaborative Deep Learning With Membership Privacy," IEEE Transactions on Dependable and Secure Computing, Vol. 20, No.3, pp. 2641-2656, 2023(DOI:10.1109/TDSC.2022.3192326)
  16. H.-K. Ko, "Privacy Preserving Techniques for Deep Learning in Multi-Party System," The Journal of the Convergence on Culture Technology (JCCT), Vol. 9, No.3, pp. 647-654, 2023(DOI: http://dx.doi.org/10.17703/JCCT.2023.9.3.647)
  17. K. KIM, S.-H. Lee, "Development of CNN-Transformer Hybrid Model for Odor Analysis," International Journal of Advanced Culture Technology, Vol. 11, No.3, pp. 297-301, 2023(DOI:https://doi.org/10.17703/IJACT.2023.11.3.297)