# Overfitting for minority class after SMOTE w/ random forests

I used SMOTE to make a predictive model, with class 1 having 1800 samples and 35000+ of class 0 samples. Hence, as per SMOTE, synthetic samples were created and the random forest was trained.

However, I am now getting most results as class 1 when I test my model. I just tried to test it on the training set and this is what I got:

Without SMOTE

With SMOTE

I've also tried hyperparameter optimisation, but that hasn't worked

Thanks

PS: Used SMOTE implementation in pandas with UnbalancedDataset library

• "I am now getting most results as class 1 when I test my model." Why is this surprising? What is the question exactly? You are biasing your model to favor class 1 by creating more observations of that class, so it obviously favors that class. Was this not what you wanted? Have I missed something? Jun 14 '16 at 13:22