基于生成对抗网络的文本序列数据集脱敏

张煜; 吕锡香; 邹宇聪; 李一戈

doi:10.11959/j.issn.2096-109x.2020046

您当前的位置：

首页 >

文章列表页 >

基于生成对抗网络的文本序列数据集脱敏

学术论文 | 更新时间：2024-06-03

- 基于生成对抗网络的文本序列数据集脱敏
- Differentially private sequence generative adversarial networks for data privacy masking
- 网络与信息安全学报 2020年6卷第4期页码：109-119
- 作者机构：
- 作者简介：
  
  [ "张煜（1995- ），男，陕西延安人，西安电子科技大学硕士生，主要研究方向为隐私保护和机器学习" ]
  [ "吕锡香（1978- ），女，陕西洛南人，西安电子科技大学教授、博士生导师，主要研究方向为网络与协议安全、机器学习与安全、密码算法与协议" ]
  [ "邹宇聪（1999- ），男，湖南桃江人，主要研究方向为隐私保护和机器学习" ]
  [ "李一戈（1995- ），男，陕西洛南人，西安电子科技大学博士生，主要研究方向为机器学习与安全" ]
- 基金信息：
  
  信息保障重点实验室基金;The Foundation of Science and Technology on Information Assurance Laboratory(KJ-17-108);陕西省重点研发计划;The Key Research and Development Project of Shaanxi Province,China(2019ZDLGY12-08);国家重点研发计划;The National Key R＆D Program of China(2018YFB0804105)
- DOI：10.11959/j.issn.2096-109x.2020046
  中图分类号： TP309.2
- 网络首发：2020-08，
  
  纸质出版：2020-08-15
- 稿件说明：
移动端阅览
张煜, 吕锡香, 邹宇聪, 等. 基于生成对抗网络的文本序列数据集脱敏[J]. 网络与信息安全学报, 2020,6(4):109-119.

Yu ZHANG, Xixiang LYU, Yucong ZOU, et al. Differentially private sequence generative adversarial networks for data privacy masking[J]. Chinese Journal of Network and Information Security, 2020, 6(4): 109-119.
张煜, 吕锡香, 邹宇聪, 等. 基于生成对抗网络的文本序列数据集脱敏[J]. 网络与信息安全学报, 2020,6(4):109-119. DOI： 10.11959/j.issn.2096-109x.2020046.

Yu ZHANG, Xixiang LYU, Yucong ZOU, et al. Differentially private sequence generative adversarial networks for data privacy masking[J]. Chinese Journal of Network and Information Security, 2020, 6(4): 109-119. DOI： 10.11959/j.issn.2096-109x.2020046.

摘要

基于生成对抗网络和差分隐私提出一种文本序列数据集脱敏模型，即差分隐私文本序列生成网络（DP-SeqGAN）。DP-SeqGAN通过生成对抗网络自动提取数据集的重要特征并生成与原数据分布接近的新数据集，基于差分隐私对模型做随机加扰以提高生成数据集的隐私性，并进一步降低鉴别器过拟合。DP-SeqGAN 具有直观通用性，无须对具体数据集设计针对性脱敏规则和对模型做适应性调整。实验表明，数据集经DP-SeqGAN脱敏后其隐私性和可用性明显提升，成员推断攻击成功率明显降低。

Abstract

Based on generative adversary networks and the differential privacy mechanism

a differentially private sequence generative adversarial net (DP-SeqGAN) was proposed

with which the privacy of text sequence data sets can be filtered out.DP-SeqGAN can be used to automatically extract important features of a data set and then generate a new data set which was close to the original one in terms of data distributions.Based on differential privacy

randomness is introduced to the model

which improves the privacy of the generated data set and further reduces the over fitting of the discriminator.The proposed DP-SeqGAN was universal

so there is no need to adjust the model adaptively for datasets or design complex masking rules against dataset characters.The experiments show that the privacy and usability of a sequence data set are both improved significantly after it is processed by the DP-SeqGAN model

and DP-SeqGAN can greatly reduce the success rate of member inference attacks against the generated data set.

关键词

Keywords

references

SWEENEY L . K-anonymity:a model for protecting privacy [J ] . International Journal of Uncertainty,Fuzziness and KnowledgeBased Systems , 2002 , 10 ( 5 ): 557 - 570 .

MACHANAVAJJHALA A , KIFER D , GEHRKE J , et al . L-diversity:privacy beyond k-anonymity [J ] . ACM Transactions on Knowledge Discovery from Data (TKDD) , 2007 , 1 ( 1 ):3.

LI N , LI T , VENKATASUBRAMANIAN S . T-closeness:privacy beyond k-anonymity and l-diversity [C ] // IEEE 23rd International Conference on Data Engineering . 2007 : 106 - 115 .

TRIASTCYN A , FALTINGS B . Generating artificial data for private deep learning [C ] // Proceedings of the PAL:Privacy-Enhancing Artificial Intelligence and Language Technologies,AAAI Spring Symposium Series . 2019 .

GILAD-BACHRACH R , DOWLIN N , LAINE K , et al . CryptoNets:applying neural networks to encrypted data with high throughput and accuracy [C ] // International Conference on Machine Learning . 2016 : 201 - 210 .

HESAMIFARD E , TAKABI H , GHASEMI M . Cryptodl:deep neural networks over encrypted data [J ] . arXiv preprint arXiv:1711.05189 , 2017

SANYAL A , KUSNER M , GASCON A , et al . TAPAS:tricks to accelerate (encrypted) prediction as a service [C ] // International Conference on Machine Learning . 2018 : 4490 - 4499 .

BOURSE F , MINELLI M , MINIHOLD M , et al . Fast homomorphic evaluation of deep discretized neural networks [J ] . IACR Cryptology ePrint Archive , 2017

MOHASSEL P , ZHANG Y . SecureML:a system for scalable privacy-preserving machine learning [C ] // 2017 IEEE Symposium on Security and Privacy (SP) . 2017 : 19 - 38 .

GRAEPEL , THORE , KRISTIN , et al . Ml confidential:machine learning on encrypted data [C ] // Information Security and Cryptology–ICISC 2012 . 2012 : 1 - 21 .

ASLETT L J M , ESPERANÇA P M , HOLMES C . A review of homomorphic encryption and software tools for encrypted statistical machine learning [J ] . Stat,2015 , 1050 :26.

宋蕾 , 马春光 , 段广晗 . 机器学习安全及隐私保护研究进展 [J ] . 网络与信息安全学报 , 2018 , 4 ( 8 ): 1 - 11 .

SONG L , MA C G , DUAN G H . Machine learning security and privacy:a survey [J ] . Chinese Journal of Network and Information Security , 2018 , 4 ( 8 ): 1 - 11 .

PARK N , MOHAMMADI M , GORDE K , et al . Data synthesis based on generative adversarial networks [J ] . Proceedings of the VLDB Endowment , 2018 , 11 ( 10 ): 1071 - 1083 .

REED S , AKATA Z , YAN X , et al . Generative adversarial text to image synthesis [C ] // International Conference on Machine Learning . 2016 : 1060 - 1069 .

YU L , ZHANG W , WANG J , et al . SeqGAN:sequence generative adversarial nets with policy gradient [C ] // Thirty-First AAAI Conference on Artificial Intelligence . 2017 .

MAYILVELKUMAR P , KARTHIKEYAN M . L-diversity on k-anonymity with external database for improving privacy preserving data publishing [J ] . International Journal of Computer Applications , 2012 , 54 ( 14 ): 7 - 13 .

WANG Q , XU Z W , QU S Z , et al . An enhanced k-anonymity model against homogeneity attack [J ] . Journal of Software , 2011 : 1945 - 1952 .

BOS J W , LAUTER K , LOFTUS J , et al . Improved security for a ring-based fully homomorphic encryption scheme [C ] // IMA International Conference on Cryptography and Coding . 2013 : 45 - 64 .

DWORK C , . Differential privacy:a survey of results [C ] // International Conference on Theory and Applications of Models of Computation . 2008 : 1 - 19 .

PHAN N , WANG Y , WU X T , et al . Differential privacy preservation for deep auto-encoders:an application of human behavior prediction [C ] // AAAI Conference on Artificial Intelligence . 2016 .

ABADI M , CHU A , GOODFELLOW I , et al . Deep learning with differential privacy [C ] // Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security . 2016 : 308 - 318 .

CHAUDHURI K , MONTELEONI C . Privacy-preserving logistic regression [C ] // Advances in Neural Information Processing Systems . 2009 : 289 - 296 .

PAPERNOT N , ABADI M , ERLINGSSON U , et al . Semi-supervised knowledge transfer for deep learning from private training data [J ] . Stat,2017 , 1050 :3.

WANG J , ZHANG J G , BAO W D , et al . Not just privacy:improving performance of private deep learning in mobile cloud [C ] // Proceedings of ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) . 2018 .

GOODFELLOW I , . Generative adversarial nets [C ] // NIPS . 2014 : 2672 - 2680 .

ARJOVSKY M , CHINTALA S , BOTTOU L . Wasserstein generative adversarial networks [C ] // Proceedings of the 34th International Conference on Machine Learning 70 . 2017 : 214 - 223 .

RADFORD A , METZ L , CHINTALA S . Unsupervised representation learning with deep convolutional generative adversarial networks [J ] . arXiv preprint arXiv:1511.06434 , 2015

CHEN X , DUAN Y , HOUTHOOFT R , et al . InfoGAN:interpretable representation learning by information maximizing generative adversarial nets [C ] // Proceedings of the 2016Neural Information Processing Systems of Information Technology IMEC . 2016 : 2172 - 2180

LEDIG C , THEIS L , HUSZÁR F , et al . Photo-realistic single image super-resolution using a generative adversarial net-work [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . 2017 : 4681 - 4690 .

ZHU J Y , KRÄHENBÜHL P , SHECHTMAN E , et al . Generative visual manipulation on the natural image manifold [C ] // European Conference on Computer Vision . 2016 : 597 - 613 .

ISOLA P , ZHU J Y , ZHOU T , et al . Image-to-image translation with conditional adversarial networks [C ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . 2017 : 1125 - 1134 .

KIM Y , . Convolutional neural networks for sentence classifica-tion [C ] // Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) . 2014 : 1746 - 1751 .

HOCHREITER S , URGEN SCHMIDHUBER J , ELVEZIA C . Long short-term memory [J ] . Neural Computation , 1997 , 9 ( 8 ): 1735 - 1780 .

MIRZA M , OSINDERO S . Conditional generative adversarial Nets [J ] . arXiv preprint arXiv:1411.1784 , 2014

LI Y , PAN Q , WANG S , et al . A generative model for category text generation [J ] . Information Sciences , 2018 , 450 : 301 - 315 .

SHOKRI R , STRONATI M , SONG C , et al . Membership inference attacks against machine learning models [C ] // 2017 IEEE Symposium on Security and Privacy (SP) . 2017 : 3 - 18 .

YouTube spam collection data set [EB ] . 2007 .

MIRONOV , RÉNYI , . Differential privacy [C ] // 2017 IEEE 30th Computer Security Foundations Symposium (CSF) . 2017 : 263 - 275 .

CARLINI N , LIU C , ERLINGSSON Ú , et al . The secret sharer:evaluating and testing unintended memorization in neural networks [C ] // 28th USENIX Security Symposium . 2019 : 267 - 284 .

DWORK C , FELDMAN V , HARDT M , et al . The reusable holdout:preserving validity in adaptive data analysis [J ] . Science , 2015 , 349 ( 6248 ): 636 - 638 .

浏览量

869

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于生成对抗网络的隐私增强联邦学习方案

差分隐私保护下的数据分级发布机制

DP-FedDRC：缓解维度坍塌的差分隐私个性化联邦学习

ISEM：基于生成多对抗网络的图像隐写增强模型

FLWD:基于联邦学习的Webshell检测方法