Integrating an Advanced Classifier in WEKA

In these days WEKA has become one of the most important data mining and machine learning tools. Despite the fact that it incorporates many algorithms, on the classification area there are still some unimplemented features. In this paper we cover some of the missing features that may be useful to researchers and developers when working with decision tree classifiers. The rest of the paper presents the design of a package compatible with the WEKA Package Manager, which is now under development. The functionalities provided by the tool include instance loading, successor/predecessor computation and an alternative visualization feature of an enhanced decision tree, using the J48 algorithm. The paper presents how a new data mining/machine learning classification algorithm can be adapted to be used integrated in the workbench of WEKA.