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.

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

Learning Objectives

1. Compare and contrast the core mechanisms of supervised, unsupervised, and reinforcement learning paradigms.
2. Analyze real-world scenarios to identify appropriate machine learning approaches and justify the choice.
3. Design a simple supervised learning model using a provided dataset and evaluate its performance.
4. Explain the role of data in training machine learning models, distinguishing between labeled and unlabeled data.
5. Classify common applications of machine learning, such as spam detection or customer segmentation, into their respective learning paradigms.

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.

Prepare & details

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.

Setup: Groups at tables with problem materials

Materials: Problem packet, Role cards (facilitator, recorder, timekeeper, reporter), Problem-solving protocol sheet, Solution evaluation rubric

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

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.

Prepare & details

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.

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

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.

Prepare & details

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.

Setup: Inner circle of 4-6 chairs, outer circle surrounding them

Materials: Discussion prompt or essential question, Observation notes template

• Analyze
• Evaluate
• Social Awareness
• Self-Awareness

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.

Prepare & details

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.

Setup: Wall space or tables arranged around room perimeter

Materials: Large paper/poster boards, Markers, Sticky notes for feedback

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

Teaching This Topic

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.

What to Expect

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

Common Misconception

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

What to Teach Instead

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.

Common Misconception

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

What to Teach Instead

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.

Common Misconception

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

What to Teach Instead

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.

Assessment Ideas

Quick Check

After the Sorting by Learning Paradigm activity, present students with the three scenarios and ask them to write which learning paradigm fits each. Collect responses to identify any lingering confusion about the distinctions between supervised, unsupervised, and reinforcement learning.

Discussion Prompt

During the Fishbowl Discussion on data bias, use the prompt: 'Imagine you are training a model to predict house prices. What kind of data would you need? Would this be supervised or unsupervised learning? Explain your reasoning.' Listen for students to connect data quality to model reliability.

Exit Ticket

After the Orange ML Confusion Matrix Gallery Walk, have students define one key vocabulary term (e.g., precision, recall) in their own words and provide one application example not discussed in class, using their notes from the walk.

Extensions & Scaffolding

• Challenge early finishers to design an adversarial example that fools their Teachable Machine model and document their process.
• Scaffolding for struggling students: Provide a partially labeled dataset and a simplified confusion matrix template to focus on interpreting results rather than building from scratch.
• Deeper exploration: Have students research a real-world ML system (e.g., medical diagnosis tools) and trace its training data lineage using public documentation.

Key Vocabulary

Machine Learning	A field of artificial intelligence where computer systems learn from data to improve performance on a task without being explicitly programmed for every step.
Supervised Learning	A type of machine learning where models are trained on labeled datasets, meaning each data point has a known correct output or category.
Unsupervised Learning	A type of machine learning where models are trained on unlabeled datasets, seeking to find patterns, structures, or relationships within the data on their own.
Reinforcement Learning	A type of machine learning where an agent learns to make a sequence of decisions by trying to maximize a reward signal through trial and error in an environment.
Training Data	The dataset used to teach a machine learning model; the model learns patterns and relationships from this data.