Computer Science · 12th Grade

Active learning ideas

Fundamentals of Machine Learning: Supervised Learning

Supervised learning can feel abstract to students, but active learning turns the theoretical into the tangible. Labs, discussions, and comparisons help students see how labeled data guides a model’s decisions, making the ‘supervised’ process visible and meaningful.

• Common Core State Standards
• CSTA: 3B-AP-09
• CSTA: 3B-DA-06

• 15–45 min
• Pairs → Whole Class
• 4 activities

Activity 01

• Simulation Game
• 45 min · Pairs

Hands-On Lab: Train Your First Classifier

Students use Google's Teachable Machine or a simple scikit-learn notebook to train an image or text classifier on a dataset they collect themselves. They deliberately include mislabeled examples and observe how this degrades accuracy. The lab closes with each pair reporting their accuracy and one insight about what made their training data better or worse.

How does a machine learning model differ from a traditional rule-based program?

Facilitation Tip

During Train Your First Classifier, circulate to ensure students are labeling data correctly before training, as errors here propagate through the entire process.

What to look for

Provide students with two scenarios: one describing predicting house prices and another describing identifying images of cats or dogs. Ask them to identify which scenario is a classification task and which is a regression task, and to briefly explain why.

• Apply
• Analyze
• Evaluate
• Create
• Social Awareness
• Decision-Making

Activity 02

• Think-Pair-Share
• 15 min · Pairs

Think-Pair-Share: Classification or Regression?

Present eight real-world prediction problems and ask pairs to categorize each as classification or regression and justify the choice. Include ambiguous cases like predicting customer satisfaction (score 1-10 versus positive/negative). Whole-class discussion reveals that the distinction sometimes depends on how you frame the business problem, not just the data.

Differentiate between classification and regression tasks in supervised learning.

Facilitation Tip

For Classification or Regression?, give students one minute to jot down their individual thoughts before pairing, ensuring all voices contribute to the discussion.

What to look for

Present students with a small, pre-labeled dataset (e.g., fruit type and color). Ask them to verbally explain how they would use this data to train a model to identify new fruits, focusing on the role of the labels.

• Understand
• Apply
• Analyze
• Self-Awareness
• Relationship Skills

Activity 03

• Socratic Seminar
• 25 min · Whole Class

Socratic Seminar: What Does 'Learning' Mean?

Open with the question: 'Is a model that scores 99% accuracy on training data but 60% on new data actually learning?' Students draw on their lab experience to discuss generalization, memorization, and the purpose of the train/test split. The teacher facilitates without providing answers, letting student reasoning drive the conversation toward overfitting.

Explain the process of training and evaluating a supervised learning model.

Facilitation Tip

In Socratic Seminar, step back after posing a question and let silence linger to give students space to formulate responses based on the readings and labs.

What to look for

Pose the question: 'Why is it crucial to evaluate a supervised learning model on data it has not seen during training?' Facilitate a discussion where students explain the concept of overfitting and the importance of the test set for assessing real-world performance.

• Analyze
• Evaluate
• Create
• Social Awareness
• Relationship Skills

Activity 04

• Gallery Walk
• 20 min · Small Groups

Gallery Walk: Algorithm Comparison

Post four posters around the room, linear regression, decision trees, k-nearest neighbors, and naive Bayes, each with a brief description, a sample use case, and a blank section labeled 'when this would struggle.' Groups rotate, add sticky notes to the struggle section, then rotate again to critique and extend each other's entries.

How does a machine learning model differ from a traditional rule-based program?

Facilitation Tip

During Gallery Walk, set a two-minute timer per station so students stay on task and absorb comparisons efficiently.

What to look for

Provide students with two scenarios: one describing predicting house prices and another describing identifying images of cats or dogs. Ask them to identify which scenario is a classification task and which is a regression task, and to briefly explain why.

• Understand
• Apply
• Analyze
• Create
• Relationship Skills
• Social Awareness

A few notes on teaching this unit

Focus on concrete examples and hands-on work rather than abstract formulas. Research shows students grasp the role of labels best when they create or curate their own datasets. Avoid diving too deep into algorithm mechanics early on; prioritize understanding the data-model relationship first. Use frequent low-stakes checks to catch misconceptions before they solidify.

Students will leave able to explain how labeled data trains a model, distinguish between classification and regression tasks, and articulate why evaluation on unseen data matters. They will also recognize common pitfalls like overfitting and misinterpreting model capabilities.

---

Watch Out for These Misconceptions

• During Train Your First Classifier, watch for students assuming more training examples automatically improve the model.

  Interrupt the lab to have students deliberately introduce errors into a portion of the training labels. After retraining, ask them to compare performance on a held-out test set and observe how data quality, not just quantity, drives results.

• During Train Your First Classifier, watch for students equating high training accuracy with a good model.

  Have students train on the full dataset and then test on the same data, noting the near-perfect accuracy. Next, give them a separate test set and ask them to explain why performance drops, linking this to overfitting.

• During Socratic Seminar, watch for students attributing understanding or meaning to the model’s decisions.

  Use the discussion to contrast the model’s statistical patterns with human understanding. Ask students to describe what the model ‘knows’ about the data versus what a human would know, grounding the conversation in their lab results.

---

Methods used in this brief

• Simulation Game
• Think-Pair-Share
• Socratic Seminar
• Gallery Walk