Issues and Challenges for the Detection of Inference Attacks

Paul Lachat (1, 2) , Nadia Bennani (1) , Veronika Rehn-Sonigo (3) , Lionel Brunie (1) , Harald Kosch (2)

1 LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
2 DIMIS - Distributed and Multimedia Information Systems
3 FEMTO-ST - Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174)

Nowadays applications produce and manage data of individual among which some may be sensitive and must be protected. Moreover, with the advent of smart applications, sensor data are produced by IoT devices in a huge quantity and sent to servers in the vicinity to be stored and processed. Meanwhile, newly discovered inference channels involving sensor data gives insights on personal data and raises new threats on individuals privacy. They escape the vigilance of traditional inference detection systems devoted to protecting personal data stored locally in a database. In this paper, we motivate the need of a distributed inference detection system acting in a general multi-database context and we highlight the issues that such a system would face.

Mots clés: Multi-database; Inference attack detection systems; Sensor data; Personal data

Domaines: Base de données [cs.DB]; Calcul parallèle, distribué et partagé [cs.DC]

Paul Lachat, Nadia Bennani, Veronika Rehn-Sonigo, Lionel Brunie, Harald Kosch. Issues and Challenges for the Detection of Inference Attacks. [Research Report] LIRIS; DIMIS; FEMTO-ST. 2022. ⟨hal-03623026v1⟩