Shielding PII to Prevent Re-identification and Preserve Utility

Summary: Models textual PII anonymization as an intractable Stackelberg game between re-identification attackers and protectors. SHIELD combines logical/ML PII detection with pluggable, attack-aware iterative anonymization, achieving constant-factor utility-loss approximation and improved privacy–utility trade-offs. (summarized by gpt-5.6-luna on Jul 26 2026)

Paper ID: 7482
Venue: SIGMOD
Year: 2026
Pagerank: 5.093636e-05
Overall Rank: 10,288 | 29.42%
DOI: 10.1145/3802109

Incoming Non-self Citations Over Time

No non-self incoming citations found for this paper in this database.

Authors

1. Shuhao Liu (Shenzhen University)
2. Wenfei Fan (Shenzhen University; University of Edinburgh)
3. Yijia Xu (South China University of Technology)

BibTeX Citation

@inproceedings{liu_sigmod26,
        title = {{Shielding PII to Prevent Re-identification and Preserve Utility}},
        author = {Liu, Shuhao and Fan, Wenfei and Xu, Yijia},
        series = {{SIGMOD} '26},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3802109},
        url = {https://dl.acm.org/doi/10.1145/3802109},
        year = {2026}
}

Incoming Citations (Sorted by Pagerank)

Showing 0 of 0 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank

Outgoing Citations (Sorted by Pagerank)

Showing 4 of 4 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
33	Consistent Query Answers in Inconsistent Databases	1999	PODS	0.00049907763
618	A Hybrid Approach to Functional Dependency Discovery	2016	SIGMOD	0.00015711835
6,403	Parallel Discrepancy Detection and Incremental Detection	2021	VLDB	5.8859374e-05
10,324	Outliers: The Good, the Bad and the Ugly	2026	SIGMOD	5.093636e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	4,699	Fast Data Anonymization with Low Information Loss	2007	VLDB
2	3,016	Plausible Deniability for Privacy-Preserving Data Synthesis	2017	VLDB
3	7,841	Quantifying identifiability to choose and audit epsilon in differentially private deep learning	2021	VLDB
4	1,746	The Boundary Between Privacy and Utility in Data Publishing	2007	VLDB
5	1,326	Resisting Structural Re-identification in Anonymized Social Networks	2008	VLDB
6	1,295	Minimality Attack in Privacy Preserving Data Publishing	2007	VLDB
7	4,037	Privacy Skyline: Privacy with Multidimensional Adversarial Knowledge	2007	VLDB
8	2,129	Data Synthesis based on Generative Adversarial Networks	2018	VLDB
9	4,913	Privacy Preserving Schema and Data Matching	2007	SIGMOD
10	3,232	Privacy-preserving Anonymization of Set-valued Data	2008	VLDB