Problem with unbalanced classes

by: weltenrand, 7 years ago

Last edited: 7 years ago

Hello. I am working on a data set with 5 classes and one class takes up over half of the whole data set, and the SVM I trained is predicting every single data point in the test set to that majority class. How can I solve this problem? Thanks in advance.



You must be logged in to post. Please login or register an account.



You can either duplicate the other lesser data til its balanced, or you can balance the data by throwing out the often-occurring data.

Try both and see which offers superior results.

-Harrison 7 years ago

You must be logged in to post. Please login or register an account.