🤖 AI in Education

How AI Works & The Role of Machine Learning

Big Idea:
AI algorithms reflect and distort human patterns in educational data.

Essential Question:
How do machine learning algorithms mirror human patterns—and where do they distort them?

🔍 Today’s Learning Objectives

  • Explain core machine learning concepts
  • Analyze how AI systems learn from educational data
  • Identify how AI can reflect or distort patterns
  • Evaluate the impact of algorithm design in schools

⏳ Today’s Agenda

  1. Opening Pattern Challenge
  2. Machine Learning Lecture + Demos
  3. Hands-On Zog Activities
  4. ML Application Design Project
  5. Algorithm Analysis (Formative Check)
  6. Reflection & Preview

🧩 Pattern Challenge

Can you identify the pattern?

Discuss:

What made some patterns easy or hard to spot?

Jigsaw Activity

Debrief

💡 Intro to Machine Learning

Key Terms:

  • Supervised Learning
  • Unsupervised Learning
  • Reinforcement Learning

ML ≠ Magic
It's pattern recognition — with flaws.

🔄 How Machines Learn from Data

Supervised:
Labeled examples → prediction

Unsupervised:
No labels → groupings & patterns

Reinforcement:
Trial and error → rewards

🔬 In Education, ML is Used To:

  • Grade writing
  • Recommend practice problems
  • Detect disengagement
  • Personalize learning paths

⚠️ But... patterns ≠ understanding

🔄 Activity Time: Zog Learning Simulation

Three Paths:

  • 🟩 Supervised: Zogling Classification
  • 🟨 Unsupervised: Alien Creature Clustering

Goal: Teach Zog using examples — then reflect
📄 Use the worksheet to document your process

🧠 Debrief Discussion

  • What type of learning did your Zog activity use?
  • How did Zog respond to data?
  • When did it go wrong?
  • What does this tell us about real AI systems?

☕ Break Time (10 minutes)

🛠️ Applied Learning Project

🧾 Use the Graphic Organizer

  1. Choose a domain (e.g., writing, math, reading)
  2. Decide what data you'd use
  3. Identify patterns to learn
  4. Pick an ML approach (S/U/R)
  5. Consider possible bias

✏️ Educational Domains (Choose One)

  • Writing Assessment
  • Math Problem-Solving
  • Language Learning
  • Reading Comprehension
  • Science Labs
  • Participation & Engagement
  • Homework Patterns
  • Formative Quiz Generation
  • Motivation & Dropout Risk
  • Personalized Learning

📣 Group Presentations

🎙 Share your educational AI design
💬 Focus on:

  • Data choices
  • ML approach
  • Challenges & biases

🧪 Formative Assessment

Algorithm Output Comparison

  • Two models, same data — different results
  • In pairs, analyze:
  • What did each model “see”?
  • What decisions might this lead to?
  • Which one reflects sound pedagogy?

🧠 Closing Reflection

  • What’s one thing about ML that clicked today?
  • What still confuses you?