Decision Tree Produces All Same Predictions and Probabilities
Hi,
I have created a decision tree model that is highly accurate at predicting a nominal binary (yes/no) outcome for both the yeses and nos.
When I write the model and then read it into a new scoring process, every prediction is exactly the same in terms of predicted outcome and probabilities of yes or no. I have experienced the same issue with these data when using various types of tree algorithms. Using naive bayes and some other models does not produce the same problem.
Could I send you the xml as well as the training and scoring files to see if you can help me figure out what is going on?
Thanks!
I have created a decision tree model that is highly accurate at predicting a nominal binary (yes/no) outcome for both the yeses and nos.
When I write the model and then read it into a new scoring process, every prediction is exactly the same in terms of predicted outcome and probabilities of yes or no. I have experienced the same issue with these data when using various types of tree algorithms. Using naive bayes and some other models does not produce the same problem.
Could I send you the xml as well as the training and scoring files to see if you can help me figure out what is going on?
Thanks!
Tagged:
0
Answers
have you checked if your tree is a real tree and not just a stump?
Best,
Martin
Dortmund, Germany
I have checked and confirmed that it is a real tree. However, I believe that I solved the problem, which turned out to be a data-related issue and unrelated to the software.
Thanks for your help!