JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD
JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD B.Tech IV MACHINE LEARNING Previous Question Paper (May - 2019 - R15)
Note: This question paper contains two parts A and B. Part A is compulsory which carries 25 marks. Answer all questions in Part A. Part B consists of 5 Units. Answer any one full question from each unit. Each question carries 10 marks.
Part A
Define active queries.
Answer
Active queries refer to a process in active learning where the learner is allowed to choose the data points from which it learns, rather than relying on a fixed, pre-labeled training set. The model selects the most informative instances to be labeled by an oracle to improve performance with fewer training examples.
List down the important objectives of machine learning.
Answer
- To develop algorithms that improve their performance automatically through experience.
- To identify patterns and structures in large datasets for predictive modeling.
- To minimize generalization error and prevent overfitting on unseen data.
What is Artificial Neural Network?
Answer
Sign in to view the complete answer paper.
List the issues in Decision Tree Learning.
Answer
Sign in to view the complete answer paper.
Define cross validation.
Answer
Sign in to view the complete answer paper.
Define Posterior Probability.
Answer
Sign in to view the complete answer paper.
Define case based reasoning.
Answer
Sign in to view the complete answer paper.
What are the merits and demerits of Lazy learners?
Answer
Sign in to view the complete answer paper.
Define genetic algorithm.
Answer
Sign in to view the complete answer paper.
List out the search methods for induction.
Answer
Sign in to view the complete answer paper.
Part B
With an example explain candidate elimination algorithm using the concept of version spaces.
Answer
Sign in to view the complete answer paper.
With a neat diagram explain the steps involved in designing a model for machine learning.
Answer
Sign in to view the complete answer paper.
Consider the following set of training examples:
| Classification | A1 | A2 |
|---|---|---|
| + | T | T |
| + | T | T |
| - | T | F |
| + | F | F |
| - | F | T |
| - | F | T |
What is the entropy of this collection of training examples with respect to the target function classification?
Answer
Sign in to view the complete answer paper.
What is the information gain of A2 relative to these training examples?
Answer
Sign in to view the complete answer paper.
Under what conditions the perceptron rule fails and it becomes necessary to apply the delta rule.
Answer
Sign in to view the complete answer paper.
Explain Bayesian belief network and conditional independence with example.
Answer
Sign in to view the complete answer paper.
Explain how hidden markov model is used to perform pattern classification?
Answer
Sign in to view the complete answer paper.
With the help of numerical example explain K nearest neighbor algorithm in detail.
Answer
Sign in to view the complete answer paper.
Compare and contrast between Lazy and Eager learners? Give the situations in which these learners are applied?
Answer
Sign in to view the complete answer paper.
Explain the process involved in reducing the complexity using the KBANN algorithm.
Answer
Sign in to view the complete answer paper.
Suppose that we want to solve the problem of finding out what a good car is by using genetic algorithms. Suppose further that the solution to the problem can be represented by a decision tree. What is the appropriate chromosome design for the given problem? Which Genetic Algorithm parameters need to be defined? What would be the suitable values of those parameters for the given problem? Provide a short explanation for each.
Answer
Sign in to view the complete answer paper.