A4.4 Ethical considerations questions for IB Computer Science
90 questions on Ethical considerations, 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 | 22 | 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.
DermaScan is a company building a smartphone app that checks photos of skin moles for signs of skin cancer. It has 250 000 photos that dermatologists have labelled as benign or suspicious. It plans to use deep learning.
- (a)
Outline one characteristic of deep learning.
2 marks
- (b)
Explain why this project is an example of supervised learning even though it uses deep learning.
2 marks
- (c)
The company must decide where to train the model and where to run it once trained.
Describe suitable hardware for:
training the model
running the model for users.
4 marks
- (d)
DermaScan intends to sell the app directly to the public without a doctor being involved.
Discuss whether DermaScan should do this.
6 marks
A city council is testing self-driving shuttle buses on a fixed route through the town centre. Each bus uses cameras and other sensors to detect pedestrians, cyclists and traffic signals, and decides how to steer and brake.
- (a)
Identify two real-world applications of machine learning that the shuttle uses.
2 marks
- (b)
The object detection model was trained on millions of labelled street images. Outline why deep learning is suited to this task.
2 marks
- (c)
Describe the hardware each bus needs to run its models while driving.
4 marks
- (d)
Outline one reason why the bus still needs a network connection.
2 marks
- (e)
A councillor says the council should be held accountable if a shuttle injures a pedestrian. Discuss who should be held responsible when a self-driving vehicle causes an accident.
4 marks
A credit card company trains a model to detect fraudulent payments. While cleaning the data, an analyst finds a small number of payments that are hundreds of times larger than the customer's usual spending.
- (a)
Define the term outlier.
1 mark
- (b)
Explain why the analyst should not simply delete these payments from the training data.
3 marks
- (c)
Outline one way the analyst could handle outliers that are found to be data entry errors.
2 marks
A research team is training a model to help diagnose a rare childhood disease. It has records for only 240 patients, and 70 of them are missing at least one blood-test value. A team member proposes deleting every incomplete record.
Discuss the team member's proposal.
6 marks
A logistics company receives about 4000 applications a year for warehouse supervisor roles. It plans to rank new applicants with a supervised machine learning model trained on the CVs of everyone who applied over the past ten years, each labelled "hired" or "not hired".
- (a)
Outline why this model is an example of supervised learning.
2 marks
- (b)
Over the past ten years, 85% of the supervisors hired were men. Explain how this could lead the model to treat female applicants unfairly.
3 marks
- (c)
The company deletes the gender field from the training data. Explain why the model could still be biased against female applicants.
3 marks
- (d)
Suggest two ways the company could reduce bias in the model.
2 marks
- (e)
The company wants the model to reject the lowest-ranked 60% of applicants automatically, without any human reading their applications. Discuss whether the company should do this.
5 marks
A city police force wants to use a machine learning system to predict which neighbourhoods are most likely to have burglaries in the coming week. The model is trained on ten years of the force's own arrest records, and officers would be sent to patrol the highest-risk areas.
- (a)
Identify two ethical issues raised by this system.
2 marks
- (b)
Explain how a feedback loop could make any bias in the system worse over time.
3 marks
- (c)
Outline one way the police force could make the system more transparent to residents.
2 marks
- (d)
Evaluate the use of this system by the police force.
5 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 fundamentals
- A4.2Data preprocessingHL only
- A4.3Machine learning approachesHL only
- A4.4Ethical considerationsThis page
Check your answers in Augment
A free account opens all 90 questions on Ethical considerations with their markschemes, and keeps a record of what you have marked.