How can I identify data leakage in my model?

How can I identify data leakage in my model?

Data leakage occurs when information from outside the training dataset is used to create the model. To identify it, check for features that shouldn't be available during prediction.

What are some common signs of data leakage?

Common signs include unusually high model performance on the training set, features that are derived from the target variable, or using future data in training.

How can I prevent data leakage?

To prevent data leakage, ensure proper data splitting, avoid using features that are not available at prediction time, and validate your model with a separate test set.

Can you give me an example of data leakage?

Sure! An example is using a feature like 'last purchase date' to predict customer churn, as it may directly correlate with the target variable.