Supervised and unsupervised data mining with an evolutionary algorithm

This paper describes our current research with RAGA (Rule Acquisition with a Genetic Algorithm). RAGA is a genetic algorithm and genetic programming hybrid that is designed for the tasks of supervised and certain types of unsupervised data mining. Since its initial release we have improved its predictive accuracy and data coverage, as well as its ability to generate more scalable rule hierarchies. These enhancements and several experiments are described.