Imperial researchers' research work, QueryCheetah, introduces an efficient, automated approach to enhance privacy in query-based systems.

Imperial’s Department of Computing has achieved another milestone in privacy and data security, earning the Distinguished Paper Award at the highly esteemed ACM Conference on Computer and Communications Security (CCS). The awarded paper, titled QueryCheetah: Fast Automated Discovery of Attribute Inference Attacks Against Query-Based Systems, introduces a transformative method for identifying privacy vulnerabilities in data-sharing systems. 

Authored by Bozhidar Stevanoski, Ana-Maria Cretu, and Professor Yves-Alexandre de Montjoye, the paper addresses the pressing need for rapid, automated solutions to identify and prevent attribute inference attacks on query-based systems (QBSs). These systems allow analysts to retrieve aggregate data from protected datasets, often in fields where privacy and data security are paramount, such as healthcare and finance. However, as systems grow increasingly complex, traditional methods of privacy testing become both labor-intensive and computationally demanding, impeding the ability to detect vulnerabilities in real time. 

QueryCheetah offers a novel solution to these challenges. This tool not only accelerates the discovery of privacy risks but also enhances the strength of potential attacks it identifies, making it 18 times faster than existing automated methods. The tool enables system developers to comprehensively evaluate the privacy landscape of their systems, even accounting for various attacker profiles and specific target scenarios. QueryCheetah’s flexible design allows it to identify attacks across diverse syntaxes and navigate around ad-hoc defenses, demonstrating its utility in real-world applications. 

“This award acknowledges not only the innovative work of our researchers but also highlights the need for advanced tools that can keep up with the fast-evolving landscape of data privacy,” commented Professor de Montjoye. “QueryCheetah represents a significant leap forward in ensuring that query-based systems can truly safeguard sensitive information in practical settings.” 

The award at ACM CCS solidifies Imperial’s Department of Computing as a leader in the field of privacy and security research. QueryCheetah is expected to make a significant impact on how organizations across sectors secure their data-sharing systems against increasingly sophisticated privacy threats.