Computer Science · 11th Grade

Active learning ideas

Machine Learning Fundamentals

Machine learning concepts stick when students experience the gap between human understanding and statistical pattern recognition firsthand. Active learning lets students move from abstract definitions to concrete evidence by building, testing, and critiquing models themselves.

• Common Core State Standards
• CSTA: 3B-AP-09
• CSTA: 3B-DA-07

• 20–50 min
• Pairs → Whole Class
• 4 activities

Activity 01

• Collaborative Problem-Solving
• 45 min · Pairs

Collaborative Problem-Solving: Teachable Machine Classifier

Students use Google's Teachable Machine to train an image classifier on three categories of their choosing. They then test the model's accuracy by varying the quantity and diversity of training examples. Pairs document what changed when they added more varied training data and present their findings.

Explain the fundamental concept of machine learning and how machines 'learn' from data.

Facilitation Tip

During the Teachable Machine Lab, circulate and ask each pair to explain one decision their model made that surprised them.

What to look for

Present students with three scenarios: 1) Identifying cats vs. dogs in images, 2) Grouping customers into segments based on purchasing behavior, 3) Training a robot to navigate a maze. Ask students to write which learning paradigm (supervised, unsupervised, reinforcement) best fits each scenario and why.

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

Activity 02

• Think-Pair-Share
• 20 min · Pairs

Think-Pair-Share: Sorting by Learning Paradigm

Present five real-world scenarios (detecting fraud, recommending music, training a chess engine, grouping news articles, predicting house prices). Students individually sort them into supervised, unsupervised, or reinforcement learning, then compare with a partner and resolve disagreements by explaining their reasoning before reporting out.

Differentiate between supervised, unsupervised, and reinforcement learning paradigms.

Facilitation Tip

For the Sorting by Learning Paradigm activity, provide only the first three sorting cards to each pair before revealing the remaining five to build gradual complexity.

What to look for

Facilitate a class discussion using the prompt: 'Imagine you are building a system to predict house prices. What kind of data would you need? Would this be supervised or unsupervised learning? Explain your reasoning.'

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

Activity 03

• Fishbowl Discussion
• 35 min · Small Groups

Fishbowl Discussion: Data Bias in ML

A small group in the center discusses a case study of a biased ML model -- such as facial recognition misidentifying darker skin tones -- while the outer ring listens. The outer ring then rotates in to add analysis, and all groups must connect the observed bias back to specific training data choices.

Analyze real-world problems that can be solved using machine learning.

Facilitation Tip

In the Fishbowl Discussion, assign the first two discussants in advance to model respectful turn-taking for their peers.

What to look for

On an index card, have students define one key vocabulary term in their own words and provide one example of its application that was not discussed in class.

• Analyze
• Evaluate
• Social Awareness
• Self-Awareness

Activity 04

• Gallery Walk
• 50 min · Small Groups

Gallery Walk: Orange ML Confusion Matrix

Groups use the Orange visual ML tool to run classification on a provided dataset and produce a confusion matrix. Each group posts their results and methodology on the wall; other groups rotate and leave one observation and one question on sticky notes before groups reconvene to respond.

Explain the fundamental concept of machine learning and how machines 'learn' from data.

Facilitation Tip

During the Gallery Walk, place a large sticky note at each station for students to post questions they have after reviewing the confusion matrices.

What to look for

Present students with three scenarios: 1) Identifying cats vs. dogs in images, 2) Grouping customers into segments based on purchasing behavior, 3) Training a robot to navigate a maze. Ask students to write which learning paradigm (supervised, unsupervised, reinforcement) best fits each scenario and why.

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

A few notes on teaching this unit

Teach machine learning by starting with familiar tools students already trust—spreadsheets and simple classifiers. Avoid overwhelming students with deep math at first; focus on data, patterns, and consequences. Research shows that students grasp abstract concepts better when they first see a model fail due to bad data, then fix it themselves.

Successful learning looks like students confidently distinguishing between supervised and unsupervised learning after hands-on experiments. You’ll see them questioning data quality and recognizing limitations of models they’ve trained, not just repeating definitions.

---

Watch Out for These Misconceptions

• During the Teachable Machine Lab, watch for students attributing human-like understanding to their model when it performs well on familiar examples.

  Pause the lab after the first successful training round and ask each pair to generate a new image that their model will likely misclassify, such as a cat drawn sideways or a dog wearing sunglasses. Have them explain why the model failed, reinforcing that it only recognizes statistical patterns, not meaning.

• During the Teachable Machine Lab, listen for students assuming that adding more images will always improve accuracy.

  After students train their first model, introduce a second dataset with many images but only two clear categories (e.g., red apples vs. green apples). Have them compare performance and discuss whether quantity alone was sufficient, highlighting the importance of representative and diverse data.

• During the Sorting by Learning Paradigm activity, watch for students equating any computer decision-making with machine learning.

  As students sort the cards, ask them to argue why a rule-based recommendation system (e.g., "if price > $50, recommend premium") belongs outside the ML category. Use this moment to clarify that ML requires learning from data, not hard-coded rules.

---

Methods used in this brief

• Collaborative Problem-Solving
• Think-Pair-Share
• Fishbowl Discussion
• Gallery Walk