Paper 1 AI and ML Drills

These are original topic-local Paper 1-style drills, not a complete 100-mark Paper 1. For this 2027 module, question style is inferred from general Computing assessment patterns rather than from past AI/ML papers.

Detailed answers are in Paper 1 AI and ML Answers.

Revise the topic hub first:

Questions

Question 1: AI Definition

Define Artificial Intelligence in the context of computing. [2]

Question 2: AI Task Examples

Give four examples of tasks that can be performed well by AI systems. [4]

Question 3: ML Versus Traditional Programming

A programmer wants to classify emails as spam or not spam.

Explain one difference between solving this task using traditional programming and solving it using machine learning. [3]

Question 4: Supervised or Unsupervised

For each scenario, state whether supervised or unsupervised learning is more suitable. Give a reason.

ScenarioLearning type and reason
A set of past images is labelled as cat or dog; a model should classify new images.
A shop has customer spending records but no customer group labels; it wants to find groups of similar customers.

[4]

Question 5: k-NN

A k-NN classifier stores labelled examples.

Explain how it predicts the label of a new data item. [4]

Question 6: k-Means

State the purpose of k-means and describe the two repeated steps used after initial centres have been chosen. [4]

Question 7: Meaning of k

Explain the meaning of in k-nearest neighbours and in k-means. [2]

Question 8: ML Workflow

Arrange these ML workflow steps in the standard ML workflow order used in this topic:

evaluate model
prepare data
make predictions
gather data
choose model
train model
tune parameters

[3]

Question 9: Problem Formulation

A school wants to predict whether a student is likely to submit homework late. It has records containing number of previous late submissions, attendance percentage, and whether the next homework was late.

Identify:

  1. two possible input features;
  2. the label;
  3. whether this is supervised or unsupervised learning.

[4]

Question 10: Model Evaluation

An ML model makes these predictions:

ItemPredictedActual
1spamspam
2not spamspam
3spamspam
4not spamnot spam

Calculate the accuracy. [2]

Review Checklist

After attempting these questions, check whether you can:

  • define AI and ML accurately;
  • give concrete AI task examples;
  • distinguish traditional programming from ML;
  • identify supervised and unsupervised learning scenarios;
  • explain k-NN and k-means without confusing their use of ;
  • order the ML workflow;
  • calculate simple accuracy.