As far as I know, it is not possible with the Weka implementation MultilayeredPerceptron OR Rapid Miner NN implementation to predict/classify more than one output (multiple class label attributes). Did anyone come across or use a Neural Network Implementation the support Multi Class Label?. Any help would be appreciated.
I have got the feeling that this is not what you are looking for, but do you know the Stuttgart Neural Network Simulator.
greetings,
Steffen