Summary: Trading off Granularity against Complexity in
Predictive Models for Complex Domains
Ingrid Zukerman, David W. Albrecht, Ann E. Nicholson and Krystyna Doktor
School of Computer Science and Software Engineering
Monash University, Clayton, VICTORIA 3800, AUSTRALIA
fingrid,dwa,annn,krysg@csse.monash.edu.au
Abstract. The automated prediction of a user's interests and requirements is an
area of interest to the Artificial Intelligence community. However, current pre­
dictive statistical approaches are subject to theoretical and practical limitations
which restrict their ability to make useful predictions in domains such as the
WWW and computer games that have vast numbers of values for variables of
interest. In this paper, we describe an automated abstraction technique which ad­
dresses this problem in the context of Dynamic Bayesian Networks. We compare
the performance and computational requirements of fine­grained models built
with precise variable values with the performance and requirements of a coarse­
grained model built with abstracted values. Our results indicate that complex,
coarse­grained models offer performance and computational advantages com­
pared to simpler, fine­grained models.
1 Introduction
It has long been recognized in the Artificial Intelligence community that problem refor­

Source: Albrecht, David - Caulfield School of Information Technology, Monash University

Collections: Computer Technologies and Information Sciences