Realization of a Knowledge Inference Rule Induction System
WU Jin-pei · Systems Engineering · 2003
One of the major obstacles in performing rule induction from empirical data is the inconsistency of information about the object. For inductive learning, the paper makes an approach to integrates respective traits of rough sets, genetic algorithms and Boolean logic. Using such an approach, a system that can discovers certain rules or possible rules form inconsistent information,has been developed. The system is validated by case data and the results of validation are presented in the paper.