Machine learning training data is the focus of this Grade 10 Computer Science worksheet on artificial intelligence. Students examine how supervised learning uses examples and labels to learn patterns, how predictions are compared with correct answers during training, and why the quality of a dataset can influence how well a model performs on new inputs.

Connect training data to model behavior

The lesson follows the supervised learning process from labeled examples through prediction, error measurement, parameter updates, and evaluation. Students distinguish training data, labels, features, predictions, epochs, evaluation data, and generalization while reasoning about how each part contributes to learning.

• Explain how labeled training examples help a model learn patterns.
• Identify how dataset size, variety, incorrect labels, and bias can affect predictions.
• Describe how comparing a prediction with a label provides feedback during training.
• Distinguish data used for training from data used to evaluate generalization.
• Interpret why a model may perform well on training examples but less accurately on new data.

Practice with realistic machine learning decisions

Students work through a dialogue question, matching activity, cloze passage, labeled diagram, and explanatory reading passage before applying the concepts through multiple-choice questions and true-or-false statements. Short-answer prompts ask students to explain training updates, the purpose of unseen evaluation data, and the effect of dataset variety.

The extended response places students in a practical email-classification scenario. They analyze why strong training accuracy might not transfer to new emails and explain how better training-data and evaluation-data practices could improve results. This makes the resource useful for checking whether students can reason about model performance rather than simply memorize AI vocabulary.

Teacher support included

A separate Teacher Guide PDF accompanies the worksheet with learning objectives, teaching tips, discussion questions, and common mistakes. These notes can support whole-class explanation or follow-up discussion when students confuse examples with labels, assume more training data is automatically better, or overlook the role of evaluation data.

Classroom use

Use this machine learning worksheet after introducing supervised learning, as guided practice in an artificial intelligence unit, for independent review, or as a written concept check. The combination of vocabulary, model-training logic, data-quality reasoning, and generalization scenarios helps students connect how a model learns with why its predictions succeed or fail on new inputs.