Supervised Learning: Classification and Regression

Active learning works for this topic because students need concrete experience to grasp the difference between classification and regression. Handling real datasets—even small ones—builds intuition about when to predict categories versus numbers, and why the problem framing matters as much as the algorithm choice.

• 11th Grade
• Computer Science
• 4 activities
• 20 min–50 min

Learning Objectives

1. Compare the predictive accuracy of a Decision Tree model versus a Linear Regression model on a given dataset.
2. Explain the fundamental difference between classification and regression tasks using concrete examples.
3. Construct a simple supervised learning model (e.g., Decision Tree or Linear Regression) to predict outcomes from a provided dataset.
4. Analyze the decision boundaries or regression line generated by a chosen algorithm to understand how it makes predictions.

Problem-Based Learning

⏱ 20 min·Pairs

Card Sort: Classification vs. Regression

Give pairs a set of problem cards (predict tomorrow's high temperature, diagnose a tumor as benign or malignant, estimate a car's resale value, classify a review as positive or negative). Partners sort them into classification or regression and write a one-sentence justification for each. Debrief addresses any cards that prompted disagreement.

Prepare & details

Explain the difference between classification and regression tasks in supervised learning.

Facilitation Tip: During the Card Sort, give each pair a timer and require them to justify their pairing of scenarios aloud before moving to the next card.

Setup: Groups at tables with access to research materials

Materials: Problem scenario document, KWL chart or inquiry framework, Resource library, Solution presentation template

• Analyze
• Evaluate
• Create
• Decision-Making
• Self-Management
• Relationship Skills

Problem-Based Learning

⏱ 35 min·Small Groups

Decision Tree Construction Activity

Provide groups with a small labeled dataset (e.g., 20 animals with features like size, diet, habitat) and ask them to build a decision tree by hand, choosing splits that best separate the classes. Groups compare their trees and discuss which features they chose and why. Connect to how algorithms like ID3 make these choices systematically.

Prepare & details

Analyze how algorithms like Decision Trees or Linear Regression make predictions.

Facilitation Tip: When constructing decision trees, have students start with a pencil and large paper so they can physically draw splits and revise without undoing digital work.

Setup: Groups at tables with access to research materials

Materials: Problem scenario document, KWL chart or inquiry framework, Resource library, Solution presentation template

• Analyze
• Evaluate
• Create
• Decision-Making
• Self-Management
• Relationship Skills

Think-Pair-Share

⏱ 20 min·Pairs

Think-Pair-Share: When Does the Algorithm Fail?

Show students an example where a decision tree overfits a small dataset (perfect accuracy on training, poor on test). Ask partners to explain in their own words what went wrong and propose one fix. Share explanations with the class. This activity surfaces overfitting intuition before formally defining it.

Prepare & details

Construct a simple supervised learning model using a given dataset.

Facilitation Tip: For the Think-Pair-Share on failure modes, assign the 'fail' role to one partner so both students actively look for weaknesses rather than only successes.

Setup: Standard classroom seating; students turn to a neighbor

Materials: Discussion prompt (projected or printed), Optional: recording sheet for pairs

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

Problem-Based Learning

⏱ 50 min·Pairs

Live Coding: Sklearn Supervised Model

Students follow along building a simple classification or regression model using scikit-learn on a provided dataset (e.g., iris flowers or housing prices). At three points, the instructor pauses and students predict what the next line of output will be before it runs. Pairs discuss predictions, then see the result. Debrief covers what the metrics mean.

Prepare & details

Explain the difference between classification and regression tasks in supervised learning.

Facilitation Tip: During the live coding session, pause after every few lines of code and ask students to predict what the next line will do before revealing it.

Setup: Groups at tables with access to research materials

Materials: Problem scenario document, KWL chart or inquiry framework, Resource library, Solution presentation template

• Analyze
• Evaluate
• Create
• Decision-Making
• Self-Management
• Relationship Skills

Teaching This Topic

Teachers approach this topic by anchoring every concept in a tangible decision students can visualize or code. Avoid abstract lectures about bias-variance tradeoffs; instead, show how a small change in tree depth affects accuracy on a test set. Research suggests that students grasp supervised learning faster when they alternate between building models by hand and debugging code, so balance unplugged activities with live coding. Warn students that ‘accuracy’ can be misleading, and model evaluation should always include a held-out test set.

What to Expect

Successful learning looks like students confidently labeling tasks as classification or regression, constructing decision boundaries with clear rules, identifying failure modes in their own models, and writing code that trains and evaluates a supervised learner. They should also articulate why simple models often outperform complex ones on small data.

Watch Out for These Misconceptions

Common Misconception

During the Card Sort activity, watch for students who assume all numeric outcomes are regression problems without considering categorical targets disguised as numbers.

What to Teach Instead

Have them revisit the sorted cards and compare examples like ‘customer churn (yes/no)’ versus ‘daily sales in dollars’; prompt them to explain why one is classification and the other regression using the card labels.

Common Misconception

During the Decision Tree Construction Activity, watch for students who build overly deep trees to reach perfect training accuracy.

What to Teach Instead

Stop the group after 10 minutes and ask: ‘Does your tree make sense to a human?’ Have them prune the tree to three splits and evaluate on a small test set to see performance drop, then discuss overfitting.

Common Misconception

During the Think-Pair-Share discussion on failure modes, watch for students who blame data quality rather than model choice or evaluation practices.

What to Teach Instead

Guide them back to the test set they created earlier and ask: ‘If the test error is high, is it the data or the way we measured success? What could we change—data, algorithm, or evaluation method?’

Assessment Ideas

Quick Check

After the Card Sort activity, present the three scenarios and ask students to label each as classification or regression. Collect responses on mini whiteboards and discuss the justifications as a class to surface misconceptions before moving on.

Exit Ticket

After the Decision Tree Construction Activity, collect each student’s hand-drawn tree and one sentence describing how they chose the first split. Check that they correctly identified the target variable as numeric or categorical and that their split reflects a meaningful rule.

Discussion Prompt

During the Think-Pair-Share activity, circulate and listen for students to explain why a classification model is appropriate for predicting student pass/fail and what ethical pitfalls (e.g., biased training data) might arise. Use their responses to seed the full-class discussion on model fairness.

Extensions & Scaffolding

• Challenge: Ask students to find a real-world dataset on Kaggle, train a decision tree classifier or regressor, and write a three-sentence report comparing its performance to a k-nearest neighbors model trained on the same data.
• Scaffolding: Provide a partially filled decision tree diagram with one split missing; students complete the tree and explain each node’s decision rule.
• Deeper exploration: Have students implement cross-validation from scratch in Python to see how splitting data into folds changes the estimate of model performance.

Key Vocabulary

Supervised Learning	A type of machine learning where an algorithm learns from a labeled dataset, meaning each data point has a known correct output or category.
Classification	A supervised learning task focused on predicting a discrete category or class label, such as 'spam' or 'not spam' for emails.
Regression	A supervised learning task focused on predicting a continuous numerical value, such as the price of a house or temperature.
Decision Tree	An algorithm that makes predictions by creating a tree-like structure of decisions based on feature values, splitting data at each node.
Linear Regression	An algorithm that models the relationship between a dependent variable and one or more independent variables by fitting a linear equation to the observed data.