Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 7, 2026Neural Computing and ApplicationsOpen Access

PrivCQ: Trading multi-dimensional conditional queries under personalised local differential privacy

View Full Paper
Ask AI
Bookmark
Share

Authors

MZMengxiao ZhangWLWeidong LiYLYiping Liu

Discussion

Loading...

Member takes

Overview

PrivCQ demonstrates enhanced query accuracy in multi-dimensional data while ensuring personalised local differential privacy.

Key Points

  • The aim is to develop a private data query system that enables conditional queries over multi-dimensional sensitive data.
  • Designed PrivCQ for conditional queries on multi-dimensional data.
  • Introduced multi-dimensional personalised local differential privacy (m-PLDP).
  • Developed total purchased privacy maximisation (TPPM) principle for procurement.
  • Proposed techniques: attribute fusion and aggregation conditioning for querying.
  • Validated three query mechanisms on real-world datasets.
  • Achieved m-PLDP across diverse scenarios.
  • Demonstrated significant linkage between query accuracy and m-PLDP.
  • Validated the effectiveness of the proposed techniques through empirical research.

Cite This Study

Zhang et al. (2026) studied this question.

synapsesocial.com/papers/69fbe2f2164b5133a91a24b1https://doi.org/10.1007/s00521-026-11988-2
View Full Paper
Ask AI
Bookmark
Share