A New Pruning Method for Associative Classification using Simulated Annealing Technique

Associative Classification (AC) algorithms normally produce large number of rules during "Rule Generation" step, so many of these algorithms use various pruning methods to eliminate the redundant or misleading rules, and consequently the size of the classifier will be reduced and the classification accuracy will be enhanced. In this paper we propose a new pruning method, namely PSA (Pruning based on Simulated Annealing), the new method tested against 7 data sets from UCI Machine Learning Repository, and the experimental results show that PSA method enhances the accuracy of the classifier.