You are an ML Engineer developing a model to aws video

Full Certification Question

You are an ML Engineer developing a model to predict credit card fraud for a financial institution. The dataset you have is extensive, containing millions of transactions, but it is highly imbalanced, with only a small fraction representing fraudulent transactions. To ensure your model generalizes well to unseen data, you need to split your dataset effectively into training, validation, and test sets. You must also avoid overfitting and ensure that the model's performance is accurately measured during development. Which of the following strategies should you employ to effectively utilize the training, validation, and test sets to develop a robust and reliable machine learning model, given the imbalanced nature of the dataset? (Select two)