What term describes data containing known outcome values for the attribute you wish to predict?

• A. Training
• B. Scoring
• C. Labeling
• D. Testing

Answer: (A)

The main idea here is recognizing the dataset used to teach a model in supervised learning. Data that includes the actual outcomes for the target you want to predict is what you use to train the model—these labeled examples let the model learn the relationship between inputs and the expected outputs. That’s why this data is called training data: it’s the material the model learns from.

Other terms describe different parts of the workflow. Scoring involves producing predictions or scores on data, not the data itself used for learning. Labeling is the act of assigning labels to data, which is a preprocessing or annotation step rather than the dataset used for training. Testing refers to data used to evaluate how well the model performs after training, providing an unbiased assessment rather than guiding the learning process.