IRI-9400568 Truszczynski, Miroslaw University of Kentucky $57,341 - 12 mos. Revision Problems: A Tool for Programming Knowledge Base Transformations This is the first year funding of a three-year continuing award. This research project is to develop a new logic-based approach to handle the effects of change in knowledge bases. It is domain independent, so it could be an important tool for people working in logic-based knowledge representations. It addresses important questions that any knowledge representation must address analogously. The system of revision programs significantly extends logic programming with stable model semantics. Revision programs are treated as input/output devices (drawing from the experience of DATALOG). As such, revision programs allow relation of (unknown) causes to (observable) effects. Consequently, they can be used to study abductive reasoning. The research program is to introduce and study the formalism of revision programming and to investigate its applications in belief revision and abduction.
