IB Computer Science HL · Year 2 · Lesson 53
Clustering, association rules, reinforcement learning and genetic algorithms
Machine-learning approaches differ in what they learn, how they are trained, how they are evaluated and which problems they suit.
Paper 150 minutesA4.3.4 · A4.3.5 · A4.3.6 · A4.3.7
Use Next or the arrow keys.
Today’s targets
What you need to be able to do
2 / 9
- DescribeHow clustering techniques in unsupervised learning are used to group data based on similarities in features
- DescribeHow learning techniques using the association rule are used to uncover relations between different attributes in large data sets
- DescribeHow an agent learns to make decisions by interacting with its environment in reinforcement learning
- DescribeThe application of genetic algorithms in various real-world situations
A4.3.4A4.3.5A4.3.6A4.3.7
Paper 1 lensMatch the depth of every response to the command term. Previously learned content can move quickly, but retrieval must still be accurate.
Page 2 of 9
Retrieve
Rapid Recall Deck
3 / 9
Say the answer aloud before flipping. Mark secure knowledge quickly and spend time on the gaps.
- What is clustering in unsupervised learning?: Grouping data points so items with similar features are placed together without using predefined class labels.
- What is a real-world use of clustering?: For example, segmenting customers into groups based on purchasing behaviour.
- What does association-rule learning look for?: Relationships in which the presence or value of one set of attributes/items is associated with another.
- What are state, action, reward, and policy in reinforcement learning?: State describes the current situation; an action is a choice by the agent; reward is feedback; the policy maps situations to actions.
- What is the exploration-versus-exploitation trade-off?: Exploration tries less-known actions to gain information; exploitation chooses actions currently believed to give high reward.
- What is the overall goal of a reinforcement-learning agent?: To learn behaviour that maximizes cumulative reward over time.
- What are the main stages or components of a genetic algorithm?: Population, fitness function, selection, crossover, mutation, evaluation, and termination.
- What kind of real-world problem suits a genetic algorithm?: Optimization where many candidate solutions can be evaluated but exhaustive search is impractical, such as route planning/travelling-salesperson-style problems.
Page 3 of 9
A4.3.4 + A4.3.5 · Learn
Core knowledge and application
4 / 9
A4.3.4Describe
How clustering techniques in unsupervised learning are used to group data based on similarities in features
Machine-learning approaches differ in what they learn, how they are trained, how they are evaluated and which problems they suit.
Exam moveGive a detailed, accurate account of what happens or what something is like.
Required detail 1Clustering techniques in unsupervised learning group data based on feature similarities
Required detail 2Real-world applications of clustering may include using purchasing data to segment a customer base
Explain it without notes
Describe: How clustering techniques in unsupervised learning are used to group data based on similarities in features in the context of a team comparing candidate models for prediction, clustering and automation.
- Clustering techniques in unsupervised learning group data based on feature similarities
- Real-world applications of clustering may include using purchasing data to segment a customer base
A4.3.5Describe
How learning techniques using the association rule are used to uncover relations between different attributes in large data sets
Machine-learning approaches differ in what they learn, how they are trained, how they are evaluated and which problems they suit.
Exam moveGive a detailed, accurate account of what happens or what something is like.
Required detail 1Mining techniques using the association rule and interpretation of the results for a given scenario For example, in crime analysis, the techniques may reveal that areas with high rates of vandalism also often experience incidents of theft, assisting law enforcement in predictive policing and resource allocation
Explain it without notes
Describe: How learning techniques using the association rule are used to uncover relations between different attributes in large data sets in the context of a team comparing candidate models for prediction, clustering and automation.
- Mining techniques using the association rule and interpretation of the results for a given scenario For example, in crime analysis, the techniques may reveal that areas with high rates of vandalism also often experience incidents of theft, assisting law enforcement in predictive policing and resource allocation
Page 4 of 9
A4.3.6 + A4.3.7 · Learn
Core knowledge and application
5 / 9
A4.3.6Describe
How an agent learns to make decisions by interacting with its environment in reinforcement learning
Machine-learning approaches differ in what they learn, how they are trained, how they are evaluated and which problems they suit.
Exam moveGive a detailed, accurate account of what happens or what something is like.
Required detail 1The principle of cumulative reward and the foundational concepts of agent–environment interaction, encompassing actions, states, rewards and policies
Required detail 2The exploration versus exploitation trade-off as a core concept in reinforcement learning
Explain it without notes
Describe: How an agent learns to make decisions by interacting with its environment in reinforcement learning in the context of a team comparing candidate models for prediction, clustering and automation.
- The principle of cumulative reward and the foundational concepts of agent–environment interaction, encompassing actions, states, rewards and policies
- The exploration versus exploitation trade-off as a core concept in reinforcement learning
A4.3.7Describe
The application of genetic algorithms in various real-world situations
Machine-learning approaches differ in what they learn, how they are trained, how they are evaluated and which problems they suit.
Exam moveGive a detailed, accurate account of what happens or what something is like.
Required detail 1For example: population, fitness function, selection, crossover, mutation, evaluation, termination
Required detail 2A real-world application of genetic algorithms is seen in optimization problems, such as route planning (e.g. the “travelling salesperson problem”)
Explain it without notes
Describe: The application of genetic algorithms in various real-world situations in the context of a team comparing candidate models for prediction, clustering and automation.
- For example: population, fitness function, selection, crossover, mutation, evaluation, termination
- A real-world application of genetic algorithms is seen in optimization problems, such as route planning (e.g. the “travelling salesperson problem”)
Page 5 of 9
Apply
Transfer to a new scenario
6 / 9
ScenarioA team comparing candidate models for prediction, clustering and automation needs a design or technical decision related to today’s topic. Explain what matters and why.
- Clustering techniques in unsupervised learning group data based on feature similarities
- Real-world applications of clustering may include using purchasing data to segment a customer base
- Mining techniques using the association rule and interpretation of the results for a given scenario For example, in crime analysis, the techniques may reveal that areas with high rates of vandalism also often experience incidents of theft, assisting law enforcement in predictive policing and resource allocation
- The principle of cumulative reward and the foundational concepts of agent–environment interaction, encompassing actions, states, rewards and policies
- The exploration versus exploitation trade-off as a core concept in reinforcement learning
- For example: population, fitness function, selection, crossover, mutation, evaluation, termination
- A real-world application of genetic algorithms is seen in optimization problems, such as route planning (e.g. the “travelling salesperson problem”)
Page 6 of 9
Exam lens
Paper 1 practice
7 / 9
Build the response before checking notesUse precise terminology and match the required depth.
- Describe: How clustering techniques in unsupervised learning are used to group data based on similarities in features in the context of a team comparing candidate models for prediction, clustering and automation.
- Describe: How learning techniques using the association rule are used to uncover relations between different attributes in large data sets in the context of a team comparing candidate models for prediction, clustering and automation.
- Describe: How an agent learns to make decisions by interacting with its environment in reinforcement learning in the context of a team comparing candidate models for prediction, clustering and automation.
Self-checkAnswer the exact command term. For explain, include mechanism/reason; for compare, pair criteria; for discuss/evaluate/justify, build supported reasoning and a conclusion.
Page 7 of 9
2027 Case Study
Generative AI research checkpoint
8 / 9
Build a reusable evidence bank for Paper 1 Section B. Keep claims technical, specific and supported.
Dataset curationSelecting, checking and organizing training data so that it is relevant, lawful and suitable for the intended use.
Know and apply in case-study context
ConditionalGeneration controlled by an input condition such as text, a class label, a sketch or another image.
Know and apply in case-study context
Training stabilityHow consistently a model can train toward useful behaviour without divergence, collapse or severe oscillation.
Know and apply in case-study context
Hybrid modelA system that combines different generative approaches so one model can offset limitations of another.
Deep for HL where tied to GAN/hybrid challenges
Page 8 of 9
Homework
Finish the learning cycle
9 / 9
Exam preparation — main task
Complete targeted 2027 case-study research and cumulative Paper 1/Paper 2 practice. Record evidence and technical vocabulary you can use in extended responses.
Retrieval
Repeat today’s recall deck and revisit any item marked ‘Review again’.
Designed by David Xu
© 2026 David Xu. All rights reserved.
Educators and students are welcome to use these materials for non-commercial teaching and learning with attribution. Please share the original link when possible. Reposting, redistributing modified copies, removing attribution, or commercial use requires prior permission.
End of Lesson 53.