Computer Science · 11th Grade

Active learning ideas

Unsupervised Learning: Clustering

Active learning works because clustering is fundamentally about recognizing patterns, and humans learn best by doing. Moving students through physical and computational simulations helps them internalize how clusters form and why choices like K matter, making abstract concepts tangible.

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

• 15–35 min
• Pairs → Whole Class
• 4 activities

Activity 01

• Stations Rotation
• 15 min · Whole Class

Human Clustering Activity

Post a scatterplot of 20 points on the board. Ask students to walk up and draw cluster boundaries using their judgment, no algorithm. Different students often draw different boundaries, which opens a discussion: what makes a cluster valid? Is there one right answer? This motivates why a formal algorithm with a defined criterion is useful.

Explain how unsupervised learning identifies patterns without explicit labels.

Facilitation Tip

During the Human Clustering Activity, walk around and ask students to explain their grouping criteria aloud so the whole class hears different approaches.

What to look for

Present students with a small 2D dataset (e.g., 6-8 points) and ask them to manually perform one iteration of the K-Means algorithm. They should draw the initial centroids, assign points, and calculate the new centroid locations.

• Remember
• Understand
• Apply
• Analyze
• Self-Management
• Relationship Skills

Activity 02

• Stations Rotation
• 35 min · Small Groups

K-Means Simulation by Hand

Groups receive a small 2D dataset printed on paper and three colored markers representing K=3 cluster centers placed at random. Following the algorithm's steps, assign, recalculate, repeat, they trace K-Means by hand until convergence. Groups compare final clusters and discuss how different random starts affected the result.

Analyze the purpose and mechanics of clustering algorithms like K-Means.

Facilitation Tip

For the K-Means Simulation by Hand, provide grid paper to keep centroids and points neatly organized, reducing calculation errors.

What to look for

Pose the question: 'Imagine you are a data scientist for a streaming service. How could you use clustering to improve user experience without knowing exactly what each user likes beforehand? What challenges might you face?'

• Remember
• Understand
• Apply
• Analyze
• Self-Management
• Relationship Skills

Activity 03

• Think-Pair-Share
• 20 min · Pairs

Think-Pair-Share: Choosing K

Show students clustering results for the same dataset with K=2, K=4, and K=7. Ask partners: which K seems most natural and why? How would you decide? After sharing, introduce the elbow method as a more systematic approach. Discuss why choosing K is a judgment call, not a formula.

Differentiate between supervised and unsupervised learning applications.

Facilitation Tip

In the Think-Pair-Share: Choosing K, give each pair a whiteboard to sketch their ideas so they can easily share their reasoning with the class.

What to look for

Ask students to write down one key difference between supervised and unsupervised learning and provide a real-world example of where clustering is applied, explaining briefly why it's unsupervised.

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

Activity 04

• Case Study Analysis
• 30 min · Small Groups

Case Study Analysis: Real Clustering Applications

Provide three short case studies: customer segmentation for a retailer, grouping news articles by topic, and detecting anomalies in network traffic. Groups identify what data was likely clustered, what features mattered, and what a business or analyst would do with the cluster assignments. Each group presents to the class.

Explain how unsupervised learning identifies patterns without explicit labels.

Facilitation Tip

During the Case Study Analysis, assign each group a different application domain so the discussion covers a broad range of uses.

What to look for

Present students with a small 2D dataset (e.g., 6-8 points) and ask them to manually perform one iteration of the K-Means algorithm. They should draw the initial centroids, assign points, and calculate the new centroid locations.

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

A few notes on teaching this unit

Teaching unsupervised learning requires balancing hands-on exploration with clear explanations of assumptions. Avoid rushing through the math; let students experience the frustration of local optima when centroids jump around. Emphasize that clustering is exploratory, not predictive, and that evaluation is always contextual. Research shows students grasp centroid movement better when they physically move objects or draw arrows on paper.

Students will explain how clusters emerge from unlabeled data, justify their choice of K, and evaluate clustering outcomes critically. They will connect mathematical steps to real-world decisions, showing they understand both the mechanics and the limitations of the algorithm.

---

Watch Out for These Misconceptions

• During Human Clustering Activity, watch for students assuming that the number of groups they identify must be the 'correct' number for the algorithm.

  Use the activity to highlight that different grouping criteria lead to different numbers of clusters. After students form groups, ask them to explain why they chose their number and then run the same activity with a different rule to show variability.

• During K-Means Simulation by Hand, watch for students believing that the algorithm will always produce the same clusters every time.

  Have students start with the same data but different initial centroids. Compare the final clusters to demonstrate sensitivity to initialization. Ask them to discuss why this matters for real-world applications.

• During Think-Pair-Share: Choosing K, watch for students thinking K must be known perfectly before running any analysis.

  Use the activity to show that K is a hypothesis. Give pairs multiple datasets and ask them to propose K values, then explain their reasoning using the elbow method or silhouette scores they sketch by hand.

---

Methods used in this brief

• Stations Rotation
• Think-Pair-Share
• Case Study Analysis