Study Guide Instructor view

Expert Systems and Machine Learning

Teaching notes for Lesson 8.5, with controls for presenting it.

Keep this page on your own screen. Share only the lesson window. The lesson window holds no teaching notes, so sharing it cannot show them.

Presenting

  1. Open the lesson in its own window and share that window, not this one.
  2. Use Next and Previous here, or the lesson’s own, to move through the lesson. Next reveals one thing at a time: the next idea, or the answer to the question on screen, then the next step. Previous takes the last reveal back. The line below says what is on screen and what Next will show.
  3. Reveal hints and answers when the class is ready, and reset a demonstration before using it again.

The controls need JavaScript and a browser that supports BroadcastChannel. Without them, open the lesson in presentation mode and use its own controls.

The two windows talk to each other only inside this browser, on this device. Nothing is sent to a server, and no account is needed. A second device (a phone or another tablet) cannot follow the lesson window yet.

Lesson plan

Level
Intermediate
Time
About 65 minutes; the steps below add up to 71

Learners start with a plant expert who asks questions and gives advice, then act as the reasoning part of a small plant adviser before any term is given. They name the four parts of an expert system, trace several customers through the rules, and discover that no matching rule means no advice. They then see a program adapt its own stored number, which introduces machine learning, and compare the two kinds of system.

Objectives

  • Explain what is meant by artificial intelligence, and give its three main characteristics
  • Name the four components of an expert system and say what each does
  • Act as an inference engine: apply a small rule base to facts to reach a conclusion
  • Explain that machine learning is a program adapting its own processes and/or data
  • Compare an expert system with a machine-learning system and decide which a scenario describes

Before the lesson

  • Display the plant facts and the rules R1 to R5 so that learners can work with them throughout.
  • Mini whiteboards for the tracing tasks.
  • Learners should have met rule-based systems and machine learning in What Counts as AI?

Steps

  1. Introduction Opening

    On screen: the lesson’s title, module and number; Next adds its objectives and key terms, as the lesson page lists them; Next again begins the first step.

    Objectives on screen

    • Explain what is meant by artificial intelligence, and give its three main characteristics
    • Name the four components of an expert system and say what each does
    • Act as an inference engine: apply a small rule base to facts to reach a conclusion
    • Explain that machine learning is a program adapting its own processes and/or data
    • Compare an expert system with a machine-learning system and decide which a scenario describes

    Purpose: Learners start with a plant expert who asks questions and gives advice, then act as the reasoning part of a small plant adviser before any term is given. They name the four parts of an expert system, trace several customers through the rules, and discover that no matching rule means no advice. They then see a program adapt its own stored number, which introduces machine learning, and compare the two kinds of system.

  2. 1. The plant expert Think 5 min

    Purpose: Start from a human expert learners can picture, and let them list what a program would need.

    Ask

    • What would a program need to know?
    • What would it have to do with what it knows?

    Responses to expect

    • Facts and rules
    • Reason from the customer’s answers to advice

    Misconceptions to challenge

    • A learner may believe: “Experts just guess.” Respond: Ask how Mrs Okafor decides. She uses what she knows.

    Checking understanding: Learners write their list before discussion.

    Depending on the answers

    • If learners say only ‘AI’: ask what data and what steps that AI would need

    In the lesson window

    • Think it through: Mrs Okafor retires, and the garden centre wants a computer program to give the same advice. What would the program need to know, and what would it have to do with it?
  3. 2. Be the advice program Discover 10 min

    Purpose: Learners apply the facts and rules themselves before the terms are given, then watch the process step by step.

    Ask

    • Which rules did you use for the wet Cactus?
    • Why does R3 apply after R1?

    Responses to expect

    • R1 then R3
    • R1’s conclusion is a new fact

    Misconceptions to challenge

    • A learner may believe: “Only one rule is ever used.” Respond: A conclusion from one rule can be the fact another rule needs.

    On the board or with the pen

    • Write the facts, then tick each rule as it fires.

    Checking understanding: The stepped trace shows who has followed the chain.

    Depending on the answers

    • If learners start with R3: ask what must be true first

    In the lesson window

    • Think it through: Using only the facts and the rules above, what advice should the customer be given, and which rules did you use to reach it? Work through them in order.
  4. 3. The four parts of an expert system Explain 9 min

    Purpose: Name knowledge base, rule base, inference engine and user interface, and separate them in the example.

    Ask

    • Which part stores the IF … THEN rules?
    • Which part does the reasoning?

    Responses to expect

    • Rule base
    • Inference engine

    Misconceptions to challenge

    • A learner may believe: “Inference means the same here as in language models.” Respond: An inference engine applies rules written by people; no trained model is involved.

    Checking understanding: The parts table and the editing question.

    Depending on the answers

    • If learners mix knowledge base and rule base: ask whether it is a fact or an IF … THEN

    In the lesson window

    • Check your understanding: The garden centre decides that, for too much water, the advice should also say to check the pot has drainage holes. Which part of the expert system must be edited?
  5. 4. Your turn as the inference engine Apply 9 min

    Purpose: Independent tracing with four customers, including one where no rule fires.

    Ask

    • Which rule finds the problem?
    • Which rule gives the advice?

    Responses to expect

    • R2 and R4; none and R5; R1 and R3; none and none
    • R5 for a Cactus in shade

    Misconceptions to challenge

    • A learner may believe: “Any rule can fire if the plant is mentioned.” Respond: Every part of the IF must be true.

    Checking understanding: The tracing table.

    Depending on the answers

    • If learners fire R5 for a Fern in the shade: ask what light a Fern needs

    In the lesson window

    • Apply it: A customer enters: plant Cactus, soil dry, position shade. What does the plant adviser output?
  6. 5. When no rule fits Predict 7 min

    Purpose: Learners reason about why a system gives no answer and what no answer means.

    Ask

    • Why might it give no answer?
    • What could the garden centre do?

    Responses to expect

    • No rule matched
    • Add a rule

    Misconceptions to challenge

    • A learner may believe: “No answer means everything is fine.” Respond: It means only that no rule matched.

    Checking understanding: The Fern with wet soil question.

    Depending on the answers

    • If learners trust silence: ask who wrote the rules and what they left out

    In the lesson window

    • Think it through: Why might an expert system give no answer? Think about who wrote its rules and what they included. What could the garden centre do about it?
    • Apply it: A garden-centre assistant says: “No advice means the Fern is healthy.” Which reply is most accurate?
  7. 6. What makes it AI? Discover 7 min

    Purpose: Define AI and its three characteristics from what learners notice about the expert.

    Ask

    • What does Mrs Okafor do that seems intelligent?
    • Which part of the adviser reasons?

    Responses to expect

    • Knows, works out, learns
    • Inference engine

    Misconceptions to challenge

    • A learner may believe: “Only learning systems are AI.” Respond: Learning is something AI can include, not a requirement.

    Checking understanding: The characteristics question.

    Depending on the answers

    • If learners say the interface reasons: ask which part draws the conclusion

    In the lesson window

    • Think it through: What does Mrs Okafor do that makes her seem intelligent, and which of those things can the plant adviser do?
    • Check your understanding: Which part of the plant adviser shows the AI characteristic “the ability to reason”?
  8. 7. A program that changes itself Experiment 9 min

    Purpose: Learners predict and complete the table of a program changing its own stored number, which introduces machine learning.

    Ask

    • How many days does the app use each week?
    • What did the program change by itself?

    Responses to expect

    • 7, 6, 5, 5
    • Its stored number

    Misconceptions to challenge

    • A learner may believe: “The app changes the plant.” Respond: It changes its own data.

    Checking understanding: The completed table and the check.

    Depending on the answers

    • If learners miss that fine keeps the number: reread the rule for fine

    Stretch

    • What rule would make the number rise if the soil was soggy?

    In the lesson window

    • Think it through: A plant is droopy in weeks 1 and 2, and fine in weeks 3 and 4. Predict the number of days the app uses in each of the four weeks.
    • Check your understanding: In the reminder app, what does the program change by itself?
  9. 8. Expert system or machine learning? Explain 7 min

    Purpose: Compare expert systems and machine learning on where knowledge comes from and how each changes.

    Ask

    • Who supplies the knowledge in each?
    • How does each change?

    Responses to expect

    • People; data and outcomes
    • A person edits it; it adapts itself

    Misconceptions to challenge

    • A learner may believe: “Anything that gives different answers is learning.” Respond: Different answers for different inputs is not learning.

    Checking understanding: The two classification questions.

    Depending on the answers

    • If learners call the tax site machine learning: ask whether it changes its own rules

    In the lesson window

    • Check your understanding: A school’s fictional revision app records which quiz questions a pupil gets wrong, and asks those questions more often in later weeks, with no teacher changing anything. Which kind of system is the part that adjusts how often each question appears?
    • Apply it: A fictional website asks a user a series of questions about their income and gives advice using IF … THEN rules written by accountants, and shows which rules led to the advice. Which description fits best?
  10. 9. Show what you know Consolidate 8 min

    Purpose: Explain the difference in learners’ own words and decide whether a rule-only system is AI.

    Ask

    • How do the two differ in how they come to know what to do?
    • Is a system that never learns AI?

    Responses to expect

    • People write rules; the program adapts itself
    • Yes, if it uses data and rules and reasons

    Misconceptions to challenge

    • A learner may believe: “AI must learn.” Respond: Learning can be included but is not required.

    Checking understanding: The explanation and the helpline question.

    Depending on the answers

    • If learners say rule-based systems are not AI: return to the three characteristics

    In the lesson window

    • Think it through: In your own words, explain how an expert system and a machine-learning system differ in how they come to know what to do. Use an example of your own, not the plant adviser.
    • Apply it: A fictional medical helpline program stores facts about illnesses and rules such as IF fever AND rash THEN advise a doctor’s visit. A reviewer says: “This system is not AI, because it never learns.” What is the best response?
  11. Lesson recap Closing

    On screen: the lesson’s objectives again; Next adds the closing question, which asks the class to explain each one in their own words without looking back. It says nothing about what anyone has learnt: use the answers to decide what to revisit.

    Points to listen for

    • AI simulates intelligent behaviour: data and rules, reasoning, and sometimes learning.
    • An expert system has a knowledge base, rule base, inference engine and user interface.
    • With no matching rule an expert system gives no answer; a person must add one.
    • Machine learning is a program automatically adapting its own processes and/or data.

Consolidation

  • AI simulates intelligent behaviour: data and rules, reasoning, and sometimes learning.
  • An expert system has a knowledge base, rule base, inference engine and user interface.
  • With no matching rule an expert system gives no answer; a person must add one.
  • Machine learning is a program automatically adapting its own processes and/or data.