Upsampling is a method of balancing the minority and majority classes by adding additional samples to the minority classes. The most trivial example is naive resampling, which randomly adds existing samples multiple times to the final balanced data set. Another more sophisticated method is SMOTE. The advantage of synthetic data is that its generation can be manipulated so that it can be trained on unbalanced original data, but balanced synthetic data can be generated which can then be used to develop predictive models.