Hi Soumen,

Handing class imbalance has a vast literature of its own and there is no one-size fits all solution to it. You need to try different method and see which works. In case of cross-validation you can create your own balanced folds using stratified resampling.

Dear Soumen Roy,

This previous question (the answers) can help you:

https://www.researchgate.net/post/How_to_approach_imbalanced_class_sizes_for_classification_and_k-fold_cross-validation

Cross validation doesn't handle class imbalance problem. It is generally used for preventing model overftiting.

You can use different sampling techniques like Undersampling, Oversampling etc.