CAREER: Privacy Preserving Security Analytics: When Security Meets Privacy

职业：隐私保护安全分析：当安全遇到隐私时

• 批准号: 2308730
• 负责人: Yuan Hong
• 金额: $ 50万
• 依托单位: University of Connecticut
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-01-01 至 2026-09-30
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2308730&HistoricalAwards=false
• 关键词: CAREER; Privacy; Preserving; Security; Analytics

A fast-growing number of enterprises and organizations have outsourced their security analytics tasks to external managed security service providers (MSSPs) for security monitoring and threat detection. However, such cost-effective and reliable security solutions currently request their service tenants to continuously outsource their large-scale and disparate datasets. This project tackles the privacy risks in such security analytics outsourcing services with rigorous privacy guarantees. This project aims to create a new paradigm of privacy preserving data analysis to privately perform real-time anomaly detection on both structured and unstructured data (e.g., network traffic, surveillance videos, system logs, and emails). The main goal is to fundamentally advance differential privacy and secure multiparty computation in this new context of privacy preserving security analytics. To this end, we propose novel differential privacy mechanisms and secure multiparty computation protocols, explore provable privacy guarantees with theoretical studies, and deploy the privacy preserving techniques in scalable real-time systems. After addressing the fundamental challenges for mitigating privacy risks in a wide variety of data and applications while ensuring high utility and efficiency, the expected research results can be leveraged to many other online monitoring and analysis applications. This project also integrates the research and education at intersections of privacy, security and data analysis. It develops a comprehensive educational and outreach program, including cybersecurity workforce training, educational materials development and distribution, K-12 outreach, and research dissemination to broader communities.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

越来越多的企业和组织将其安全分析任务外包给外部托管安全服务提供商 (MSSP)，以进行安全监控和威胁检测。然而，这种经济高效且可靠的安全解决方案目前要求其服务租户不断外包其大规模且不同的数据集。该项目通过严格的隐私保证来解决此类安全分析外包服务中的隐私风险。该项目旨在创建一种新的隐私保护数据分析范例，以私下对结构化和非结构化数据（例如网络流量、监控视频、系统日志和电子邮件）进行实时异常检测。主要目标是在隐私保护安全分析的新背景下从根本上推进差异隐私和安全多方计算。为此，我们提出了新颖的差分隐私机制和安全多方计算协议，通过理论研究探索可证明的隐私保证，并将隐私保护技术部署在可扩展的实时系统中。在解决了减轻各种数据和应用程序中的隐私风险同时确保高实用性和效率的基本挑战之后，预期的研究结果可以用于许多其他在线监控和分析应用程序。该项目还整合了隐私、安全和数据分析交叉点的研究和教育。它制定了全面的教育和推广计划，包括网络安全劳动力培训、教材开发和分发、K-12 推广以及向更广泛社区传播研究成果。该奖项反映了 NSF 的法定使命，并通过使用基金会的评估进行评估，被认为值得支持。智力价值和更广泛的影响审查标准。

项目成果

Interpretation Attacks and Defenses on Predictive Models Using Electronic Health Records

• DOI: 10.1007/978-3-031-43418-1_27
• 发表时间: 2023
• 作者: Fereshteh Razmi;Jian Lou;Yuan Hong;Li Xiong

WPES '22: 21st Workshop on Privacy in the Electronic Society

WPES 22：第 21 届电子社会隐私研讨会

• DOI: 10.1145/3548606.3563220
• 发表时间: 2022
• 期刊: Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security
• 作者: Hong, Yuan;Wang, Lingyu
• 通讯作者: Wang, Lingyu

Text-CRS: A Generalized Certified Robustness Framework against Textual Adversarial Attacks

• DOI: 10.1109/sp54263.2024.00053
• 发表时间: 2023-07
• 期刊: ArXiv
• 作者: Xinyu Zhang;Hanbin Hong;Yuan Hong;Peng Huang;Binghui Wang;Zhongjie Ba;Kui Ren

UniAP: Protecting Speech Privacy With Non-Targeted Universal Adversarial Perturbations

• DOI: 10.1109/tdsc.2023.3242292
• 发表时间: 2024-01
• 期刊: IEEE Transactions on Dependable and Secure Computing
• 影响因子: 7.3
• 作者: Peng Cheng;Yuexin Wu;Yuan Hong;Zhongjie Ba;Feng Lin;Liwang Lu;Kui Ren

Task-Agnostic Privacy-Preserving Representation Learning for Federated Learning against Attribute Inference Attacks

用于对抗属性推断攻击的联邦学习的任务无关的隐私保护表示学习

• DOI: 10.1609/aaai.v38i10.28965
• 发表时间: 2024
• 期刊: Proceedings of the AAAI Conference on Artificial Intelligence
• 作者: Arroyo Arevalo, Caridad;Noorbakhsh, Sayedeh Leila;Dong, Yun;Hong, Yuan;Wang, Binghui
• 通讯作者: Wang, Binghui