This invited talk overviews 20 years of work at the intersection between the two AI areas of Knowledge Representation (KR) and Machine Learning (ML). The distinguishing feature of this research is the extension of the methodological apparatus of Inductive Logic Programming (ILP) along a couple of directions towards the realm of Description Logics (DLs). One aims at learning hybrid rules that tightly integrate DATALOG and DLs, whereas the other aims at learning axioms in fuzzy DLs. Both have turned out to be alternative suitable ways to treat spatial knowledge in several applications and could be successfully applied also in the field of Forensics.
Combining Knowledge Representation and Machine Learning in Forensics
Francesca Alessandra Lisi
2020-01-01
Abstract
This invited talk overviews 20 years of work at the intersection between the two AI areas of Knowledge Representation (KR) and Machine Learning (ML). The distinguishing feature of this research is the extension of the methodological apparatus of Inductive Logic Programming (ILP) along a couple of directions towards the realm of Description Logics (DLs). One aims at learning hybrid rules that tightly integrate DATALOG and DLs, whereas the other aims at learning axioms in fuzzy DLs. Both have turned out to be alternative suitable ways to treat spatial knowledge in several applications and could be successfully applied also in the field of Forensics.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
