neural network


I have some questions about neural network in Knime

First of all, i want to run a MLP with cross validation. Is it possible to do it with both numeric and categorical attributes?

If it is not the case, are there other alternatives?

Thanks for responses



Hi wafa, 

Have a look at our cross validation meta node.  It will allow you to plug in virtually any learner and predictor evaluate the robustity of your model.  If your data contains string attributes, also consider looking at the decision tree learner/predictor, it is easy to use and pretty powerful. 





Thank you Aaron