Home › AKTU PYQ › Machine Learning Techniques › Important Questions › Unit 2

Machine Learning Techniques Unit 2 – Regression, Bayesian Learning and Support Vector Machine: important questions for AKTU

Unit 2 (Regression, Bayesian Learning and Support Vector Machine) questions that AKTU repeats most often. This unit carries about 34 marks per paper. Start with the repeated questions, then the most asked topics.

Most repeated Unit 2 questions

  • Describe Linear Regression and Logistic Regression with differences. (7 marks, 2026, Logistic Regression) – also asked in 2021, 2023, 2024, 2025
  • Explain types of kernels used in Support Vector Machines. (7 marks, 2026, Types of Support Vector Kernels) – also asked in 2021, 2023, 2024
  • Explain hyperplane (decision boundary) in SVM. Categorize various popular kernels associated with SVM. (10 marks, 2023, Types of Support Vector Kernels) – also asked in 2021, 2026

Most important Unit 2 topic

  • Logistic Regression (Unit 2: Regression, Bayesian Learning and Support Vector Machine) – asked 7 times in 2021, 2023, 2024, 2025, 2026

Most asked Unit 2 topics

  • Logistic Regression – asked 7 times
  • Naive Bayes Classifier – asked 6 times
  • Concept Learning – asked 5 times

More Unit 2 previous year questions

  • State Bayes Theorem. (2 marks, 2026, Bayes Theorem)
  • State any two applications of Logistic Regression. (2 marks, 2026, Logistic Regression)
  • Define Support Vector Machine (SVM). (2 marks, 2026, Introduction to Support Vector Machine)
  • Describe Linear Regression and its assumptions. (7 marks, 2026, Linear Regression)
  • Explain the working of Naïve Bayes Classifier with assumptions. (7 marks, 2026, Naive Bayes Classifier)

Unit 2 syllabus topics

  • Linear Regression
  • Logistic Regression
  • Bayes Theorem
  • Concept Learning
  • Bayes Optimal Classifier
  • Naive Bayes Classifier
  • Bayesian Belief Networks
  • Expectation-Maximization Algorithm
  • Introduction to Support Vector Machine
  • Types of Support Vector Kernels
  • Hyperplane and Decision Surface
  • Properties of SVM