An abstract reasoning architecture for privacy policies monitoring

Privacy is a topic of increasing interest not only for scientific communities, but also for public opinion and for regulatory bodies from all Countries. The problem is that privacy management and enforcement are very dynamic activities and monitoring known fallacies of systems is really difficult, since their number grows up day by day. Moreover, monitoring systems for yet unknown vulnerabilities is, of course, even more difficult. Here we want to present a model and an architecture, based on semantics and reasoning, which is able to detect privacy problems by abstract reasoning on monitored information. The methodology we present here fetches data and evidences from logs and other files, applying model transformation techniques in order to populate semantics repository for enacting reasoning actions. We provide here some experimental results in order to prove the strength of the proposed approach. © 2020