Dear Sharare Ashtari

Yes you can increase your data points using the sampling method. Here I have attached few links that may help you out in clear understanding:

For tabular data:

1. https://www.kdnuggets.com/2017/06/7-techniques-handle-imbalanced-data.html

2. https://machinelearningmastery.com/data-sampling-methods-for-imbalanced-classification/

3. https://www.analyticsvidhya.com/blog/2020/07/10-techniques-to-deal-with-class-imbalance-in-machine-learning/

4. https://towardsdatascience.com/increasing-the-amount-and-diversity-of-data-part-1-2025caf9d53c

For Image dataset:

5. https://medium.com/@thimblot/data-augmentation-boost-your-image-dataset-with-few-lines-of-python-155c2dc1baec

6. https://towardsdatascience.com/how-to-use-data-augmentation-to-10x-your-image-datasets-dab42858be55

Sharare Ashtari You should check resampling or bootstrapping strategies which is especially useful for such cases, and you can estimate an error band to get an idea of how useful your model is when you oversample.

Yes, you can resample your data using SMOTE variants of resampling methods. Beware, your model might overfit in fewer cases.

This should not be done. It is better to use adaptive intervals on numeric scales

You can try Generative Adversarial Networks they can outperform other techniques

Check oversamplimg methods like SMOTE

https://towardsdatascience.com/5-smote-techniques-for-oversampling-your-imbalance-data-b8155bdbe2b5

Before Adapting any sampling technique, just check out number of classes. Is it a binary classification? Check out with Data Augmentation techniques.