Machine Learning Techniques Unit 3 – Decision Tree Learning and Instance-Based Learning: important questions for AKTU
Unit 3 (Decision Tree Learning and Instance-Based Learning) questions that AKTU repeats most often. This unit carries about 25 marks per paper. Start with the repeated questions, then the most asked topics.
Most repeated Unit 3 questions
- Explain the ID3 algorithm for Decision Tree learning. (7 marks, 2026, ID-3 Algorithm) – also asked in 2024, 2025
- Illustrate the steps of the ID3 algorithm, including how it selects attributes and constructs the tree. (7 marks, 2025, ID-3 Algorithm) – also asked in 2024, 2026
- Provide a detailed explanation of the ID3 algorithm used for constructing decision trees. Discuss the key steps involved in the iterative process of feature selection and node splitting. (10 marks, 2024, ID-3 Algorithm) – also asked in 2025, 2026
Most important Unit 3 topic
- Issues in Decision Tree Learning (Unit 3: Decision Tree Learning and Instance-Based Learning) – asked 7 times in 2021, 2022, 2023, 2025, 2026
Most asked Unit 3 topics
- Issues in Decision Tree Learning – asked 7 times
- k-Nearest Neighbour Learning – asked 6 times
- ID-3 Algorithm – asked 4 times
More Unit 3 previous year questions
- Explain the concept of overfitting in Decision Trees? (2 marks, 2026, Issues in Decision Tree Learning)
- What is Entropy in Decision Tree Learning? (2 marks, 2026, Entropy and Information Theory)
- Define k-Nearest Neighbor algorithm. (2 marks, 2026, k-Nearest Neighbour Learning)
- Discuss Information Gain and its role in Decision Tree Learning. (7 marks, 2026, Information Gain)
- Explain k-Nearest Neighbour (k-NN) algorithm and its limitations. (7 marks, 2026, k-Nearest Neighbour Learning)
Unit 3 syllabus topics
- Decision Tree Learning Algorithm
- Inductive Bias
- Inductive Inference with Decision Trees
- Entropy and Information Theory
- Information Gain
- ID-3 Algorithm
- Issues in Decision Tree Learning
- k-Nearest Neighbour Learning
- Locally Weighted Regression
- Radial Basis Function Networks
- Case-Based Learning