A4.1 Machine learning fundamentals questions for IB Computer Science
90 questions on Machine learning fundamentals, written by Augment in the style of IB exam papers. Six of them are below in full, to answer on paper.
- IB Computer Science
- Part of A4 Machine learning
- SL and HL
Questions by paper and level
SL and HL questions are for students at either level. HL only questions cover what only HL students study.
| Paper | SL and HL | HL only | All |
|---|---|---|---|
| Paper 1 | 68 | 20 | 88 |
| Paper 2 | 2 | 0 | 2 |
| Every paper | 70 | 20 | 90 |
Counted on
Questions to try
Written by Augment in the style of IB exams, not past papers. Answer them on paper, then check each one against its markscheme with a free account.
A hospital has 40 000 chest X-ray images. A radiologist has tagged each image as either pneumonia or healthy. The hospital wants a machine learning model that flags new X-rays that may show pneumonia.
- (a)
Identify the type of machine learning the hospital should use.
1 mark
- (b)
Outline why the radiologist's tags are needed to train this model.
2 marks
- (c)
Describe how the trained model would be used when a new patient's X-ray is taken.
3 marks
An email provider wants to stop spam messages reaching its users' inboxes. Its engineers decide to train a machine learning model on millions of emails that users have already marked as "spam" or "not spam".
- (a)
Outline how abstraction is used when each email is turned into data the model can learn from.
2 marks
- (b)
Describe how pattern recognition is central to this approach.
3 marks
- (c)
Identify one objective and one constraint for the spam filter.
2 marks
- (d)
Outline two evaluation criteria for the spam filter.
4 marks
- (e)
A manager says: "A filter that catches 99% of spam must be a success."
Explain why this may not be true.
2 marks
A supermarket chain stores every till receipt from its 120 stores. Its data team wants to find out which products customers tend to buy together, without deciding in advance what to look for.
- (a)
State the type of machine learning that suits this task.
1 mark
- (b)
Outline why this type of machine learning suits the supermarket's data.
2 marks
- (c)
Outline one way the supermarket could use the results of this market basket analysis.
2 marks
A national park has set up 60 motion-triggered cameras. The cameras have taken about 200 000 photos in a year. Many photos are empty (set off by wind moving the grass), and rangers want to know how many animals of each species visit each part of the park.
- (a)
Describe how decomposition could be applied to this problem.
3 marks
- (b)
Explain how pattern recognition can be used to identify the species in the photos.
3 marks
- (c)
Identify two items of data, other than the image itself, that should be kept with each photo.
2 marks
- (d)
Outline two evaluation criteria the rangers could use to judge the system.
4 marks
- (e)
State one constraint on the system.
1 mark
A bank is planning two machine learning projects. One will predict whether a card payment is fraudulent, using past payments that investigators have marked as fraud or genuine. The other will look for groups of customers with similar spending habits.
Distinguish between supervised learning and unsupervised learning, referring to both of the bank's projects.
4 marks
A logistics company is developing small robots that move parcels around a warehouse. Instead of programming every route, the engineers let each robot learn how to move around shelves and other robots. A robot gains points when it delivers a parcel quickly and loses points when it bumps into something.
- (a)
Identify the type of machine learning described.
1 mark
- (b)
Outline the role of the points that a robot gains and loses.
2 marks
- (c)
Outline why this type of learning does not need a labelled data set.
2 marks
- (d)
The engineers train the robots first in a computer simulation of the warehouse.
Explain one reason for training in a simulation before using the real warehouse.
3 marks
- (e)
Once trained, each robot must decide its next move many times per second.
Explain why the robot should run its model on an edge device inside the robot rather than send its sensor data to a cloud server.
3 marks
Practice all 90 in Augment
Free with an account. Every question has its markscheme, and Augment keeps the ones you got wrong so you can come back to them.
More questions in Machine learning
- A4.1Machine learning fundamentalsThis page
- A4.2Data preprocessingHL only
- A4.3Machine learning approachesHL only
- A4.4Ethical considerations
Check your answers in Augment
A free account opens all 90 questions on Machine learning fundamentals with their markschemes, and keeps a record of what you have marked.