Unit 4: A4 Machine learning
Topic 4.1: Machine learning fundamentals Questions
Practice 20 exam-style questions for IB Computer Science Topic 4.1. Review the question stems below, then unlock the full Question Bank to access markschemes, model answers, and AI grading.
1Outline3 marks
2026
Outline how supervised learning works.
Markscheme and model answer locked
Unlock Question2Outline3 marks
2026
Outline how unsupervised learning works.
Markscheme and model answer locked
Unlock Question3Identify2 marks
2026
Identify which type of machine learning suits teaching a program to play a board game well, and justify your answer.
Markscheme and model answer locked
Unlock Question4Outline2 marks
2026
Outline what is meant by training data.
Markscheme and model answer locked
Unlock Question5Explain4 marks
2026
Explain why a model's accuracy figure can be misleading.
Markscheme and model answer locked
Unlock Question6Suggest4 marks
2026
Suggest two ways the environmental cost of training could be reduced.
Markscheme and model answer locked
Unlock Question8Outline3 marks
2026
Outline how reinforcement learning works.
Markscheme and model answer locked
Unlock Question9Identify2 marks
2026
Identify which type of machine learning suits sorting customers into groups with similar buying habits, and justify your answer.
Markscheme and model answer locked
Unlock Question10Identify2 marks
2026
Identify which type of machine learning suits deciding whether an email is spam, and justify your answer.
Markscheme and model answer locked
Unlock Question11Explain3 marks
2026
Explain why training and test data must be kept separate.
Markscheme and model answer locked
Unlock Question12Outline2 marks
2026
Outline what is meant by overfitting.
Markscheme and model answer locked
Unlock Question13Explain4 marks
2026
Explain two ways overfitting can be reduced.
Markscheme and model answer locked
Unlock Question14Determine3 marks
2026
Determine the accuracy of a model that correctly classifies 456 of 600 test images, and suggest one further figure that should be reported.
Markscheme and model answer locked
Unlock Question15Suggest4 marks
2026
Suggest why access to machine learning hardware raises questions of fairness.
Markscheme and model answer locked
Unlock Question16Determine3 marks
2026
Determine what proportion of a 20 000-example dataset should typically be held back for testing, and explain the trade-off.
Markscheme and model answer locked
Unlock Question17Explain3 marks
2026
Explain why a machine learning model may be described as a black box.
Markscheme and model answer locked
Unlock Question18Suggest4 marks
2026
Suggest why a hospital might prefer a simpler, slightly less accurate model.
Markscheme and model answer locked
Unlock Question19Suggest4 marks
2026
Suggest why a model that performed well in testing may perform poorly once deployed.
Markscheme and model answer locked
Unlock Question20State2 marks
2026
State the three broad types of machine learning.
Markscheme and model answer locked
Unlock QuestionReady to practice Topic 4.1?
Get instant AI feedback on your answers, view detailed markschemes, and track your progress across all IB Computer Science topics.
Start Practising Now