Extracting trees from trained SVM models using a TREPAN based approach

This paper describes the application of a hybrid intelligent system (HIS) to extract decision trees from a trained support vector machine (SVM) model based on the TREPAN algorithm. TREPAN, a well-known technique developed originally to extract linguistic rules from a trained artificial neural network, is modified to cope with SVM models. The proposed approach is tested on five data sets with excellent performance results.