Neuro-Symbolic Artificial Intelligence and Process Mining: a Promising Marriage

Neuro-Symbolic Artificial Intelligence (NeSy AI) stems from the fields of subsymbolic AI (such as Neural Networks) and symbolic AI (Knowledge Representation and Reasoning) and tries to combine the strengths of both paradigms. In particular, the capability of Neural Networks to learn in a robust way from (noisy) examples has to be tightly coupled with the expressivity and the meaning of the symbols in a symbolic system along with its (explainable) reasoning capabilities. NeSy AI has many applications in different AI subfields and it can be a promising technique in Process Mining as well. This talk will introduce the main concepts behind NeSy AI with some examples of well-known frameworks. A discussion will follow about some features of the Process Mining field that make it suitable for applications of NeSy AI techniques. Finally, the talk will introduce some examples of Process Mining works that can be already classified as NeSy AI and can be considered as the basis for more complex and complete NeSy AI systems.

Dr. Ivan Donadello

Ivan Donadello is an Assistant Professor at Free University of Bozen-Bolzano. He received his Ph.D. degree in Computer Science from UniversitĂ  degli Studi di Trento and Fondazione Bruno Kessler in 2018 with a thesis on Neural-Symbolic AI for Semantic Image Interpretation. His current research interest mainly focuses on different subfields of Process Mining, such as, Predictive/Prescriptive Process Monitoring and tools for declarative languages. He is also interested on AI for healthcare, in particular, on virtual (dialogical) agents that support users in adhering to healthy lifestyles. His expertise encompasses the fields of Knowledge Representation (ontologies and Fuzzy Logic), Machine/Deep Learning, Computer Vision, eHealth and Explainable Artificial Intelligence.