Question - How do you avoid the overfitting of your model?
Answer -
Overfitting basically refers to a model that is set only for a small amount of data. It tends to ignore the bigger picture. Three important methods to avoid overfitting are:
- Keeping the model simple—using fewer variables and removing major amount of the noise in the training data
- Using cross-validation techniques. E.g.: k folds cross-validation
- Using regularisation techniques — like LASSO, to penalise model parameters that are more likely to cause overfitting.