You are developing a machine learning model using data stored in Google BigQuery. This dataset contains several values classified as Personally Identifiable Information (PII), such as names, addresses, and social security numbers. To comply with data privacy laws and reduce the sensitivity of the dataset before using it for training, you need to anonymize or mask these sensitive columns without removing them, as every column is critical for the model's performance. How should you proceed? | Google Professional Machine Learning Engineer Quiz