Practice A4.3 Machine learning approaches (HL only) with authentic IB Computer Science (First Exam 2027) exam questions for both SL and HL students. This question bank mirrors Paper 1, 2, 3 structure, covering key topics like programming concepts, algorithms, and data structures. Get instant solutions, detailed explanations, and build exam confidence with questions in the style of IB examiners.
A smart city initiative uses federated learning to train machine learning models across multiple hospitals, schools, and government agencies while preserving data privacy.
Explain the federated learning process and how it enables collaborative machine learning without centralizing sensitive data.
Discuss the challenges and limitations of federated learning in real-world smart city deployments.
A computer vision system uses convolutional neural networks for medical image analysis including X-rays, MRIs, and CT scans for diagnostic assistance.
Explain how convolutional layers, pooling layers, and fully connected layers work together in CNN architectures for medical image classification.
Describe the challenges of training CNNs on medical images and explain why transfer learning is particularly valuable in healthcare applications.
A healthcare AI system uses ensemble methods and deep learning to analyse medical images for cancer detection and treatment planning. Explain different ensemble learning techniques and their advantages for medical diagnosis applications:
| Ensemble Method | Combination Strategy | Diversity Source | Medical Application | Uncertainty Quantification |
|---|---|---|---|---|
| Bagging | Voting | Bootstrap sampling | - | - |
| Boosting | Sequential learning | - | Rare disease detection | - |
| Stacking | Meta-model | Different algorithms | - | High |
Discuss how uncertainty quantification in medical AI models helps clinicians make informed decisions.
Explain different ensemble learning techniques and their advantages for medical diagnosis applications:
| Ensemble Method | Combination Strategy | Diversity Source | Medical Application | Uncertainty Quantification |
|---|---|---|---|---|
| Bagging | Voting | Bootstrap sampling | - | - |
| Boosting | Sequential learning | - | Rare disease detection | - |
| Stacking | Meta-model | Different algorithms | - | High |
A cybersecurity system employs various machine learning techniques including support vector machines, random forests, and neural networks for threat detection and malware classification.
Design a multi-algorithm approach for cybersecurity threat detection:
| Algorithm Type | Input Features | Detection Capability | False Positive Rate | Computational Overhead | Adversarial Robustness |
|---|---|---|---|---|---|
| Support Vector Machines | - | Binary classification | - | - | - |
| Random Forest | - | - | Low | Medium | - |
| Neural Networks | Raw network traffic | High | - | - | - |
| Ensemble Methods | - | - | - | - | High |
Analyse how adversarial attacks specifically target machine learning security systems and describe defence strategies.
A financial services company implements reinforcement learning for algorithmic trading and portfolio optimization in dynamic market environments.
Compare different reinforcement learning algorithms for financial applications, considering sample efficiency, stability, and interpretability.
Explain how the exploration-exploitation trade-off applies to algorithmic trading and describe strategies to balance this trade-off.
A search engine company uses machine learning for query understanding, document ranking, and personalized search results across billions of web pages.
Explain how different machine learning approaches contribute to modern search engine functionality, including natural language processing and ranking algorithms.
Describe how online learning enables search engines to adapt quickly to changing user behaviour and emerging topics.
A recommendation system uses collaborative filtering and matrix factorization techniques to suggest products to customers based on purchase history and user similarities.
Compare collaborative filtering approaches (user-based vs item-based) and discuss their computational complexity and recommendation quality.
Explain how the cold start problem affects recommendation systems and describe solutions to address this challenge.
A manufacturing company uses clustering algorithms and anomaly detection to identify production defects and optimize quality control processes.
Compare different clustering algorithms for manufacturing data analysis, considering data types, scalability, and interpretability requirements.
Explain how anomaly detection techniques can identify equipment failures and quality defects in manufacturing processes.