The influence of parameters in evolutionary based rule extraction method from neural network

Abstract

In the paper the experimental study of the influence of parameters on the final results of the rule extraction method from neural network for classification problem is described. The method is based on evolutionary approach, where for each class evolves separate population. The paper starts on the presentation of the basic concepts of the method. Next, the results of experiments are described. They examine the influence of genetic parameters. Then, the parameters that affect the rule extraction efficiency are tested. All experiments are made with using UCI data sets. At the end, some general conclusion concerning the role of the parameters and their influence on the final results are formulated.