"Learning Rules From Examples in Rule-Based Systems Via an
Integer Programming Approach"
Proceedings of the Annual Conference of the Inter'l Association of Knowledge Engineers,
University of Maryland, June 26-28, pp. 403-426, 1989.
by Triantaphyllou, E., and S.R.T. Kumara
Abstract:
An increasingly interesting issue in Rule-Based System (RBS) development
is the problem of inferring rules from examples. As RBS's attempt to solve
larger and more complex problems, automatic ways for determining the knowledge bases
of the these systems become critical in Knowledge Engineering. The present paper proposes
the use of an Integer Programming approach provides a flexible and powerful way for utilizing
information that very often is difficult to handle, mainly due to the complexity of the problem.
(size = 1,445 KB)
