aimnova.
DashboardMy LearningPaper MasteryStudy Plan

Aimnova site navigation

Stay in the loop

Get the latest study resources and updates

New features, study tips and exam insights — straight to your inbox.

IB Diploma

  • IB Past Papers
  • IB Study Notes
  • IB Question Bank
  • IB Mock Exams
  • IB Revision

IB Subjects

  • IB Math AA
  • IB Math AI
  • IB Economics
  • IB Business Management
  • IB Physics
  • IB Biology
  • View all IB subjects→

IB Past Papers

  • IB Math AA HL Past Papers
  • IB Math AA SL Past Papers
  • IB Math AI HL Past Papers
  • IB Math AI SL Past Papers
  • IB Economics HL Past Papers
  • IB Economics SL Past Papers
  • IB ESS Past Papers
  • View all past papers→

Study Resources

  • Study Notes
  • Question Bank
  • Mock Exams
  • Flashcards
  • Revision Guide
  • Exam Skills
  • Command Terms
  • Grade Calculator
  • Exam Timetable 2026

Aimnova

  • Features
  • Pricing
  • For Schools
  • For Parents
  • About Us
  • Blog
  • Contact
aimnova.

AI-powered study platform for smarter revision, past-paper analysis and examiner-style feedback.

TermsPrivacyCookies·© 2026 Aimnova. All rights reserved.8afc4e3

Aimnova is not affiliated with or endorsed by the International Baccalaureate Organization (IB).

NotesComputer Science HLTopic 4.3
Unit 4 · Machine learning · Topic 4.3

IB Computer Science HL — Machine learning approaches

Machine learning approaches

Higher Level students should use this topic hub as a map: start with the shared sub-topics, then follow the HL-only extensions and exam-skill links where this topic asks for deeper analysis.

Exam technique guidePractice questions

Key concepts in Machine learning approaches

Key Idea: The approaches divide by what feedback the learner gets: the correct answers (supervised), no answers at all (unsupervised), or a score for how things turned out (reinforcement). That one question places almost every method in this topic.

Paper 1

  • Define a method; compute a measure.
  • Identify which approach a scenario needs.

Paper 2

  • Evaluate a model against an alternative.
  • Reason about a reward or fitness function.

Both

  • Name what is given up, not only what is gained.

🧭 Placing a method

FeedbackFamilyMethods here
The right answersSupervisedLinear regression · classification
NoneUnsupervisedClustering · association rules
A rewardReinforcementAgents learning from consequences
A fitness scoreSearch, not learningGenetic algorithms
"How much?" is regression — a number on a scale. "Which one?" is classification — a category. The same data supports either; the question decides.

📏 Judging a model

Never quote accuracy alone

  • Ask what a majority-class guesser would score — if it is close, accuracy is meaningless
  • Precision — of those flagged, how many really were?
  • Recall — of the real cases, how many were caught?
  • Moving the threshold trades one against the other; it is a human judgement
  • Cancer screening fears a false negative; a spam filter fears a false positive
Important: An association rule with 80% confidence is worthless if the consequent appears in 90% of baskets anyway — lift 0.89, meaning less than chance. Always divide by how common the consequent already is.

🧠 Networks

What the marks are for

  • A unit is inputs × weights, summed, plus a bias, through an activation
  • Without a non-linear activation, stacked layers collapse into one
  • Training changes weights and biases; the structure is fixed beforehand
  • A CNN adds convolution, shared weights and pooling — a feature is recognised wherever it appears
  • Neither can explain itself, which is why interpretable models still win in regulated domains

⚖️ Choosing what to deploy

ConstraintChooseBecause
A decision must be justifiedInterpretableA network has no rule to read out
Must run on a phoneSmallA large model cannot run there at all
Few hundred recordsSimpleA complex model will fit noise
Nothing else constrainedMost accurateThen accuracy is the criterion

📝 Exam-style questions

IB-style questionIdentify[4 marks]

Identify the machine learning approach for each, with a reason: (a) predicting next month's electricity demand, (b) grouping news articles with no categories defined, (c) teaching a drone to land, (d) deciding whether a transaction is fraudulent.

🔒 Model answer plan

See the mark-by-mark plan — for / against / judgement, with marking guidance — in study mode.

Claim your free topic →
IB-style questionEvaluate[6 marks]

A screening model catches 88 of 100 disease cases and wrongly flags 300 of 9,900 healthy people. A simpler, interpretable model catches 80 and wrongly flags 150. Evaluate which should be deployed.

🔒 Model answer plan

See the mark-by-mark plan — for / against / judgement, with marking guidance — in study mode.

Claim your free topic →

✅ Quick check

Cover the answers.

Which question places a method? What feedback does the learner get? Correct answers, none, or a reward.

Precision or recall — of the real cases, how many were caught? Recall. Precision is: of those flagged, how many really were?

Why is confidence alone misleading for association rules? It rewards whatever is popular. Lift divides by how common the consequent already is.

What does training a neural network change? The weights and biases only. Layers and units are hyperparameters, fixed beforehand.

When is the most accurate model the wrong choice? When a decision must be justified or audited, or when it cannot run where it is needed.


Exam tips

  • Ask what feedback the learner gets — that places almost every method.
  • "How much?" regression · "which one?" classification.
  • Never quote accuracy alone; say what a majority-class guesser would score.
  • Name which error the DOMAIN cannot afford, then move the threshold accordingly.
  • Check lift, not just confidence.
  • Say what you are giving up. A verdict with a condition attached scores highest.

What you'll learn in Topic 4.3

  • 4.3.1 Linear regression
  • 4.3.2 Classification
  • 4.3.3 Hyperparameter tuning
  • 4.3.4 Clustering
  • 4.3.5 Association rules
  • 4.3.6 Reinforcement learning
  • 4.3.7 Genetic algorithms
  • 4.3.8 Neural networks
  • 4.3.9 CNNs
  • 4.3.10 Model selection
Suggested study order: Read the notes for each sub-topic below → test yourself with flashcards → attempt practice questions → review exam technique.

Study resources — 4.3 Machine learning approaches

4.3.1

Linear regression

Notes
4.3.2

Classification

Notes
4.3.3

Hyperparameter tuning

Notes
4.3.4

Clustering

Notes
4.3.5

Association rules

Notes
4.3.6

Reinforcement learning

Notes
4.3.7

Genetic algorithms

Notes
4.3.8

Neural networks

Notes
4.3.9

CNNs

Notes
4.3.10

Model selection

Notes

Ready to study Machine learning approaches?

Get expert practice questions with instant AI feedback, and a study planner tailored to your IB Computer Science HL exam date.

Start studying free

Topic 4.3 Machine learning approaches forms a core part of Unit 4: Machine learning in IB Computer Science HL. Mastering these concepts will strengthen your understanding of connected topics across the syllabus and prepare you for exam questions that require analysis, evaluation, and real-world application.

Previous topic
4.2 Data preprocessing
Next topic
4.4 Ethical considerations
All Computer Science HL topics
Exam technique

Ready to practice?

Get AI-graded practice questions, mock exams, flashcards, and a personalised study plan — all aligned to your IB syllabus.

Start Studying Free

No credit card required · No time limit