Rule Optimization Via SG-TRUNC Method

Most inductive learning systems generate complete and consistent descriptions. In order to achieve completeness and consistency in the presence of noise or imprecision. one may generate overly complex and detailed descriptions. Such descriptions. however. may not perform well in future cases and suffer the disadvantage of excessive complexity. This is th~ well known phenomenon of overfining, In thIs paper. a rule optimization method called SQ-TRUNC is described and evaluated experimentally, SQ· TRUNC improves previous TRUNC methods and has been implemented in a more efficient way. In the method, an optimized description is obtained through a sequence of generalization and/or specialization operations performed on a complete and cons,istent concept description. The operations apphed always simplify a description. The method has been implemented in AQ16 that has been applied to two domains: a designed testing problem and "multiplexer" F 11' The experimental results have shown that both simplicity and perfonnance improvements can be gained in the domains when: noise is present.