Zhou Jiehui, Wang Xumeng, Wong Jason K, Wang Huanliang, Wang Zhongwei, Yang Xiaoyu, Yan Xiaoran, Feng Haozhe, Qu Huamin, Ying Haochao, Chen Wei
IEEE Trans Vis Comput Graph. 2023 Jan;29(1):809-819. doi: 10.1109/TVCG.2022.3209391. Epub 2022 Dec 16.
Data privacy is an essential issue in publishing data visualizations. However, it is challenging to represent multiple data patterns in privacy-preserving visualizations. The prior approaches target specific chart types or perform an anonymization model uniformly without considering the importance of data patterns in visualizations. In this paper, we propose a visual analytics approach that facilitates data custodians to generate multiple private charts while maintaining user-preferred patterns. To this end, we introduce pattern constraints to model users' preferences over data patterns in the dataset and incorporate them into the proposed Bayesian network-based Differential Privacy (DP) model PriVis. A prototype system, DPVisCreator, is developed to assist data custodians in implementing our approach. The effectiveness of our approach is demonstrated with quantitative evaluation of pattern utility under the different levels of privacy protection, case studies, and semi-structured expert interviews.
数据隐私是发布数据可视化时的一个重要问题。然而，在隐私保护可视化中表示多种数据模式具有挑战性。先前的方法针对特定的图表类型，或者在不考虑可视化中数据模式重要性的情况下统一执行匿名化模型。在本文中，我们提出了一种可视化分析方法，该方法有助于数据保管人生成多个隐私图表，同时保持用户偏好的模式。为此，我们引入模式约束来对用户对数据集中数据模式的偏好进行建模，并将其纳入所提出的基于贝叶斯网络的差分隐私（DP）模型PriVis中。开发了一个原型系统DPVisCreator，以协助数据保管人实施我们的方法。通过在不同隐私保护级别下对模式效用进行定量评估、案例研究和半结构化专家访谈，证明了我们方法的有效性。
