Which of the following techniques is used to address class imbalance in datasets?

Class imbalance occurs when the classes in a dataset are not represented equally, which can significantly affect the performance of machine learning models. To address this issue, several techniques can be employed, and all of the methods listed in the choices are effective in combating class imbalance.

Random oversampling involves duplicating instances of the minority class to balance the number of examples in each class. By replicating these instances, the overall dataset becomes more balanced, which helps the model learn the characteristics of the minority class better.

Random undersampling, on the other hand, involves removing instances from the majority class to achieve a balance between the classes. This method reduces the size of the dataset but can lead to the loss of potentially valuable information if too many instances from the majority class are removed.

The Synthetic Minority Oversampling Technique (SMOTE) is a more advanced method that generates synthetic instances of the minority class rather than simply duplicating existing ones. It creates new examples by interpolating between existing ones in the minority class, thus providing a richer and more diverse dataset which can improve model learning and generalization.

Since all these techniques aid in mitigating the impact of class imbalance, the correct answer is that all of the above options are valid approaches.