Ricardo Gonçalves, Matthias Knorr and João Leite, professors from the Department of Computer Science and researchers from the Knowledge-Based Systems Group at NOVA LINCS, won the best article at the (CORE Rank A) 17th European Conference on Logics in Artificial Intelligence (JELIA 2021) , for their work “On Syntactic Forgetting under Uniform Equivalence”, in collaboration with Tomi Janhunen (University of Tampere, Finland).
Forgetting is an operation that allows the removal, from a knowledge base, of intermediate variables that are no longer considered relevant. This study was initially triggered by the desire to eliminate (temporary) variables that represented auxiliary concepts, with the aim of restoring the declarative character of a knowledge base, or simply simplifying it. Today, it is becoming a necessary operation to properly deal with legal and privacy issues, including, for example, the application of the new EU General Data Protection Regulation, which includes the right to be forgotten. In this article, the authors build on their previous theoretical results about the limits of what can be forgotten without losing additional information and present a concrete operator that achieves the desired result in modular knowledge bases with minimal disruption.
The European Conference on Logics in Artificial Intelligence is a biennial forum for the discussion of logic-based approaches to Artificial Intelligence. The 17th edition of JELIA was held online, May 17-20, 2021, at Klagenfurt University, Austria.