VeriX-Anon: A Multi-Layered Framework for Mathematically Verifiable Outsourced Target-Driven Data Anonymization

2026-04-14

AI Summary

This study addresses the challenge faced by data owners in verifying whether cloud service providers faithfully execute agreed-upon anonymization algorithms when outsourcing privacy-sensitive data. To this end, the authors propose a three-layer verifiable k-anonymization framework that innovatively integrates three orthogonal mechanisms: Merkle hashing based on authenticated decision trees, twin records with encrypted identifiers augmented by boundary sentinels, and an interpretable AI fingerprint derived from SHAP value distributions measured via Wasserstein distance. Experimental results demonstrate that the proposed approach successfully detects 11 out of 12 attack scenarios, achieves an average F1-score improvement of 0.1574 over blind anonymization, and enables client-side verification of datasets with one million records in under one second, thereby offering a balanced solution in terms of security, data utility, and verification efficiency.

Technology Category

• Machine Learning: Privacy
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Philosophy and Ethics of AI: Accountability, Interpretability & Explainability

Application Category

• Security and Privacy: Data transparency and provenance
• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Organisations increasingly outsource privacy-sensitive data transformations to cloud providers, yet no practical mechanism lets the data owner verify that the contracted algorithm was faithfully executed. VeriX-Anon is a multi-layered verification framework for outsourced Target-Driven k-anonymization combining three orthogonal mechanisms: deterministic verification via Merkle-style hashing of an Authenticated Decision Tree, probabilistic verification via Boundary Sentinels near the Random Forest decision boundary and exact-duplicate Twins with cryptographic identifiers, and utility-based verification via Explainable AI fingerprinting that compares SHAP value distributions before and after anonymization using the Wasserstein distance. Evaluated on three cross-domain datasets against Lazy (drops 5 percent of records), Dumb (random splitting, fake hash), and Approximate (random splitting, valid hash) adversaries, VeriX-Anon correctly detected deviations in 11 of 12 scenarios. No single layer achieved this alone. The XAI layer was the only mechanism that caught the Approximate adversary, succeeding on Adult and Bank but failing on the severely imbalanced Diabetes dataset where class imbalance suppresses the SHAP signal, confirming the need for adaptive thresholding. An 11-point k-sweep showed Target-Driven anonymization preserves significantly more utility than Blind anonymization (Wilcoxon $p = 0.000977$, Cohen's $d = 1.96$, mean F1 gap $+0.1574$). Client-side verification completes under one second at one million rows. The threat model covers three empirically evaluated profiles and one theoretical profile (Informed Attacker) aware of trap embedding but unable to defeat the cryptographic salt. Sentinel evasion probability ranges from near-zero for balanced datasets to 0.52 for imbalanced ones, a limitation the twin layer compensates for in every tested scenario.

Problem

• outsourced data anonymization
• verifiable computation
• k-anonymization
• privacy-preserving
• cloud outsourcing

Innovation

• Verifiable Anonymization
• Target-Driven k-Anonymity
• Explainable AI (XAI)
• Merkle Hashing
• Boundary Sentinels