Exam AWS Certified Machine Learning - Specialty topic 1 question 310 discussion

https://aws.amazon.com/about-aws/whats-new/2022/04/amazon-sagemaker-data-wrangler-supports-random-sampling-stratified-sampling/

A. Balanced Class Representation. Stratified sampling divides the original dataset into strata (groups) based on the class labels. It then selects instances from each stratum in a proportional manner, ensuring that the class distribution in the training and validation datasets reflects the original class distribution. Improved Generalization. By having a balanced representation of all classes in the training and validation datasets, the model is exposed to a diverse range of instances during training. This helps the model learn the distinguishing features of each class more effectively, leading to better generalization performance on the validation dataset. Addressing Imbalanced Data. Stratified sampling directly addresses the issue of imbalanced data, which was identified as the root cause of the model's poor generalization performance on the validation dataset.

Stratified sampling

A: Yes - Stratified sampling ensures that each class is proportionally represented and mitigates the impact of class imbalance on model performance B: No - additional data about the majority classes does not solve class imbalance issue C: No - Does not solve class imbalance issue and may worsen the situation D: No - selecting data points at regular intervals does not solve class imbalance issue