Measuring Legislature-Aligned Privacy Risks in Synthetic Graphs

Embedding-Space Anonymization for Privacy-Preserving AI Systems

This paper studies embedding-space privacy as a representation-level learning problem. Rather than altering raw records directly, the proposed framework applies embeddingspace transformation to full-record representations through Gaussian perturbation and adversarial representation sanitization. The method is evaluated through ablation across utility metrics, linkage attacks, attribute-inference attacks, and membership-inference tests. The primary empirical evaluation uses a synthetic fusion recommendation benchmark built from MovieLens [1], [2] 32M behavior and Adult-derived demographics [3], while a secondary synthetic medical benchmark is used to examine cross-domain transferability under more constrained conditions. The strongest results appear in the recommendation experiments. Under grouped demographic privacy evaluation, the combined condition preserves recommendation utility with $N D C G {@} K=0.6312$ while reducing exact and entity linkage from 0.7090/0.7204 to 0.0001/0.0000. Sensitive-target attacker performance remains near the majority baseline, supporting the claim of empirical privacy improvement without visible ranking degradation in that benchmark. The healthcare experiments also demonstrate meaningful embedding transformation and linkage reduction, though the current benchmark remains datalimited and therefore less conclusive for utility-focused evaluation. Overall, the findings support the conclusion that embeddingspace transformation can preserve downstream utility while substantially reducing linkage risk and sensitive-information recoverability under explicit attacker evaluation. The findings support embedding-space transformation as a practical privacypreserving strategy for embedding-driven AI systems under explicit attacker evaluation.

FAIR and Square: Privacy Compliance Framework for Healthcare Databases

Toward Personalized Differentially Private Learning for Decentralized Local Graphs

Graph-structured data is increasingly generated and stored in decentralized environments, such as social platforms, mobile applications, and edge networks, where users maintain control over their local graph data. However, collecting and analyzing such decentralized graph data for downstream learning tasks raises significant privacy concerns, as nodes and their attributes often contain sensitive personal information. Local Differential Privacy (LDP) has emerged as a promising solution for privacy-preserving data collection without relying on trusted servers. Nevertheless, existing LDP-based graph learning methods typically assume uniform privacy requirements across users, ignoring the heterogeneous and personalized privacy preferences commonly observed in real-world systems. This uniform treatment leads to inflexible noise injection at the data collection stage, resulting in substantial distortion of graph data and degraded utility in subsequent analysis. To address this limitation, we propose PPGNN, a personalized differentially private framework for decentralized graph data. PPGNN enables user-specific privacy budgets during local perturbation while preserving analytical utility. To handle heterogeneous privacy levels and noise distortion, we design a two-stage solution consisting of a Personalized Perturbation Mechanism (PPM) and a weighted calibration strategy, FlexProp. Extensive experiments on six real-world graph datasets demonstrate that PPGNN effectively balances personalized privacy protection and data utility in decentralized graph learning scenarios.

Scalable privacy risk assessment of medical data processed by large language models

Dependency Triad: A Metric to Quantify the Dependencies Between Attributes for Local Differential Privacy

Collecting multidimensional user data is essential for extracting rich insights across various applications. Local Differential Privacy (LDP) has emerged as a de facto standard for mitigating privacy risks in such scenarios. A key challenge in privacy-preserving multidimensional data collection lies in inter-attribute dependencies, as they can inadvertently reveal correlated information and increase privacy vulnerabilities. Therefore, accurately measuring correlation-induced privacy leakage (CPL) is essential for privacy analysis and privacy-utility trade-off. However, existing CPL analysis solutions either require accurate prior knowledge or face scalability challenges for large numbers of attributes and high-cardinality attributes. These limit their practical applicability in real data. To address this research gap, we propose a novel metric, ``Dependency Triad''(DT), which summarizes the pairwise dependency information relevant to CPL using three parameters and yields a \emph{constant-time} conservative estimator of pairwise CPL. DT explicitly models uncertainty in prior distributional knowledge through its parameters, delivering robust leakage estimates. Moreover, its robustness to sparse distributions makes it particularly suitable for high-cardinality attributes, while the pairwise formulation serves as a tractable building block for assessing total leakage in multidimensional settings. Extensive experiments on both synthetic and real datasets demonstrate that DT consistently estimates CPL across diverse dependency regimes and prior uncertainties.

Decision Tree-Based Synthetic Data Generation Framework for Privacy-Preserving Data Publishing

The growing demand for data sharing in healthcare, scientific research, and public policy raises an ongoing challenge of protecting individual privacy while keeping data useful for analysis. This research introduces a Decision Tree-based synthetic data generation method, DTSDG and evaluates its performance using six well-known anonymization techniques. , namely, ‘k-Anonymity’, ‘l-Diversity’, ‘t-Closeness’, ‘Differential Privacy (DP)’,  ‘Generative Adversarial Networks (GANs)’, and ‘Copula-GAN’ The proposed method creates synthetic records step by step. Each attribute is predicted using a trained decision tree, which offers a balance of interpretability, efficiency, and accuracy. A composite scoring framework combines normalized Nearest Neighbor Distance (NND), Jensen–Shannon Divergence (JSD), and Wasserstein Distance into a unified metric is employed to assess how effectively each of these methods balance the conflicting performance measures: the privacy preservation and the data utility. Experiments were conducted using six diverse datasets, namely, ‘Adult’, ‘ATUS’, ‘FARS’, ‘Breast Cancer Wisconsin’, ‘cardiovascular disease’, ‘Heart Disease Cleveland’. The experimental results demonstrate that the proposed DTSDG method consistently achieves stable and competitive scores across all these datasets providing the best balance between privacy preservation and the data utility. The proposed DTSDG approach could be a practical and promising solution for publishing data while protecting privacy. This is especially true in situations where computational resources are limited or there is a strong need for model transparency.