You are working on a classification problem using time series data. After minimal experimentation, without employing sophisticated algorithms or extensive hyperparameter tuning, your model achieves an AUC ROC value of 99% on the training data. This result is surprisingly high and raises concerns about potential issues. Given the scenario, what are the two most appropriate next steps to diagnose and resolve the issue? (Choose two correct options)

Real Exam

Combat model overfitting by opting for a simpler algorithm.

1.9%

Mitigate data leakage through the application of nested cross-validation in model training.

34.6%

Counter data leakage by eliminating features that show high correlation with the target variable.

11.5%

Reduce model overfitting by adjusting hyperparameters to lower the AUC ROC score.

5.8%

Validate the model's performance on a completely unseen dataset to ensure the high AUC ROC is not due to data leakage.

46.2%

Google Professional Machine Learning Engineer