Generating rules and reasoning under inconsistencies
Guoyin Wang, Yu Wu, Feixu Liu · 2002
As the amount of information in the world steadily increases, there is a growing demand for tools for analyzing this information. In this paper, we investigate the problem of data mining, i.e. constructing decision rules from a set of primitive input data. The main contention is that there is a need to be able to generate decision rules and to reason in presence of inconsistencies. Propositional default rules are generated in this paper. Based on Skowron's default rule generation method (see T. Mollestad & A. Skowron, Proc. 9th Internat. Symposium on Foundations of Intell. Syst., pp. 448-457, 1996) and our analysis of inconsistencies, we develop a method for default rule generation from a decision table and its corresponding reasoning method. Any as-yet-unseen object can be processed with the rules generated by our rule-generating method and reasoning method.