[Moderator note - We have not, in the past, had submissions of geographically restricted interest like seminar notices. As long as the volume is moderate I will forward them to the list, they are an interesting ephemeral record of research activity. Comments on the appropriateness and/or interest in such notices welcome - mail to [EMAIL PROTECTED] and I'll summarize - tnx, Bruce] Probabilistic and Fuzzy Object-Oriented Modelling and Programming in FRIL++ BISC Seminar Tru Cao Department of Information Technology Ho Chi Minh City University of Technology and BISC Program-UC Berkeley September 20, 2001 373 Soda Hall 4:00-5:00 pm Abstract: Probabilistic and Fuzzy Object-Oriented Modelling and Programming in FRIL++ It is witnessed that object orientation has been a very useful methodology for modelling real-world problems and constructing large-scale software. In the classical object-oriented model, each object is "to be or not to be" a member of a class, and each property of a class is totally applicable to its objects. However, in the real world, it is often that both such membership and applicability are the matter of degrees, in particular ones of uncertainty. Moreover, not as in the classical object-oriented model, properties of objects in the real world may have imprecise values. We introduce a logic-based probabilistic and fuzzy object-oriented model, and propose probabilistic default reasoning on fuzzy events for uncertain property inheritance and class recognition . Our choice of the logic-based model instead of the procedure-based one is to have logical grounds for justifying the approach. The results however can be applied to development of general object-oriented systems involving uncertainty and imprecision. In particular, it provides a formal basis for the design and implementation of FRIL++, the object-oriented extension of FRIL, a logic programming language dealing with both probability and fuzziness. We present the key features of FRIL++ and its applications to user modelling and machine learning. Bio: Tru Cao is a lecturer (on leave) of the Department of Information Technology, Ho Chi Minh City University of Technology. He received his B.Eng. in Computer Science and Engineering from Ho Chi Minh City University of Technology in 1990, M.Eng. in Computer Science from Asian Institute of Technology in 1995, and Ph.D. in Computer Science from University of Queensland in 1999. For the last two years he has worked on the FRIL++ project in the Artificial Intelligence Group at University of Bristol. He is currently visiting Berkeley Initiative in Soft Computing, University of California at Berkeley, doing research on fuzzy logic applied to the Internet. His research interests are uncertain and imprecise knowledge representation and reasoning, non-classical logics and their applications, conceptual structures, object-oriented systems, and intelligent Internet. ------- End of Forwarded Message
