Reducing Number of Decision Rules by Joining

Sets of decision rules induced from data can often be very large. Such sets of rules cannot be processed efficiently. Moreover, too many rules may lead to overfitting. The number of rules can be reduced by methods like Quality-Based Filtering [1,10] returning a subset of all rules. However, such methods may produce decision models unable to match many new objects. In this paper we present a solution for reducing the number of rules by joining rules from some clusters. This leads to a smaller number of more general rules.