Re: How to evaluate and rank the contributions of the independent variables in the dependent variable in a Multilayer Perceptron Model

> Hello everyone, May I ask how we can evaluate and rank the contributions of the independent variables in the dependent variable in a Multilayer Perceptron Model. I have built a MLP model with eight independent variables and 1 dependent variable. The correlation coeffiient is good as 98%. However I don't know how to determine which variable contributed the most or which one the least in the model I have estimated. My variables are all of numerical type. Can you help please? Thanks.

Since the model won't tell you the importance of an attribute, you
could try attribute selection. Select the ClassifierAttributeEval
(with your MLP as base classifier) and the Ranker as search method.
That will output a merit score per attribute, i.e., how important a
single attribute (independent variable) was for building a model
against your class attribute (dependent variable).

Cheers, Peter
-- 
Peter Reutemann
Dept. of Computer Science
University of Waikato, NZ
+64 (7) 577-5304
http://www.cms.waikato.ac.nz/~fracpete/
http://www.data-mining.co.nz/
_______________________________________________
Wekalist mailing list -- [email protected]
Send posts to [email protected]
To unsubscribe send an email to [email protected]
To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html