Nominal-scale Evolving Connectionist Systems
Michael John Watts · The 2006 IEEE International Joint Conference on Neural Network Proceedings · 2006
A method is presented for extending the evolving connectionist system (ECoS) algorithm that allows it to explicitly represent and learn nominal-scale data without the need for an orthogonal or binary encoding scheme. Rigorous evaluation of the algorithm over benchmark data sets shows that it is able to learn, generalise and adapt well to classification problems. The algorithm is potentially useful for data mining tasks.