AKTU Machine Learning (KOE073) Question Paper 2023-24

Note: Attempt all Sections. If require any missing data; then choose suitably.

Section A

Q1. Attempt all questions in brief · 2 × 10 = 20 marks

1. What is machine learning?
2. What are the steps involved in designing learning system in machine learning?
3. Explain Artificial Neural Network.
4. What do you understand by gradient descent
5. Explain Bayes Classifier.
6. What are the basics of sampling theory?
7. What is mistake bound model of learning?
8. Explain Case based Learning.
9. How do you evaluate the performance of a model based on first-order rules?
10. What is Reinforcement Learning?

Section B

Q2. Attempt any three of the following · 10 × 3 = 30 marks

1. Explain the Differentiate between Supervised, Unsupervised and Reinforcement 10 Learning.
2. Define the following in decision tree algorithm: (i) Entropy, (ii) Information 10 gain, (iii)Gini index, (iv) Gain Ratio, (iv)Chi-Square
3. Explain Expectation Maximization algorithm.
4. Explain Backpropagation algorithm in artificial neural network (ANN) with 10 suitable example.
5. Explain the concept of hypothesis space search in the context of machine 10 learning. How does the choice of hypothesis space impact the learning process? Provide examples to illustrate the significance of hypothesis space search.

Section C

Q3

Attempt any one part of the following · 10 × 1 = 10 marks

1. Explain the concept of inductive bias in machine learning, and provide an 10 example to illustrate how it influences the learning process.
2. Explain the Candidate Elimination algorithm in machine learning and provide a 10 step-by-step solution for a hypothetical scenario.

Q4

Attempt any one part of the following · 10 × 1 = 10 marks

1. What is the difference between forward propagation and backward propagation 10 in neural networks explain weight calculation for forward pass network?
2. Consider a single-layer neural network with one input neuron, one hidden 10 neuron, and one output neuron. The activation function used is the sigmoid function. The network is trained to learn the XOR function.
   • Input layer: One neuron
   • Hidden layer: One neuron with a sigmoid activation function
   • Output layer: One neuron with a sigmoid activation function
   Input Output
   Weights and Biases (Initial Values):
   • Weight between input and hidden layer: wih 0.5
   • Bias at the hidden layer: bh =0
   • Weight between hidden and output layer: who =−0.5
   • Bias at the output layer: bo =0
   Learning Rate: 0.1
   Perform one iteration of training using the given data and update the weights and biases.

Q5

Attempt any one part of the following · 10 × 1 = 10 marks

1. Suppose we have a dataset of weather conditions and corresponding play 10 decisions (play or not play) as follows: Day Weather Temperatu Humidity Windy Play re Day 1 Sunny Hot High False No Day 2 Sunny Hot High True No Day 3 Overcast Hot High False Yes Day 4 Rain Mild High False Yes Day 5 Rain Cool Normal False Yes Day 6 Rain Cool Normal True No Day 7 Overcast Cool Normal True Yes Day 8 Sunny Mild High False No (i) If the weather is sunny, then the Player should play or not? (ii) If the level of humidity is Normal & it is windy then the player should play or not?
2. Suppose there is a rare disease that affects 1 in 10,000 people. A diagnostic test 10 for this disease has a 99% accuracy rate for both true positives (correctly identifying a person with the disease) and true negatives (correctly identifying a person without the disease). If a randomly selected person takes the test and it comes back positive, what is the probability using bayes theorem that they actually have the disease?

Q6

Attempt any one part of the following · 10 × 1 = 10 marks

1. What is mistake bound model of learning? Explain the concept, algorithm 10 evaluation, its benefit and limitation.
2. Suppose you have a dataset with the following points in a 2-dimensional space: and their corresponding labels are: A, A, B, B, B . Now, we want to classify a new point (6, 5) using k-nearest neighbors with k = 3.

Q7

Attempt any one part of the following · 10 × 1 = 10 marks

1. Consider a binary string optimization problem where the goal is to evolve a 10 binary string of length 8 to maximize the number of ones. You decide to use a simple genetic algorithm for this task. (i) Define the representation of an individual in the population. (ii) Specify the initialization step for the genetic algorithm. (iii) Outline the different operations used in genetic algorithm.
2. What are the different types of reinforcement? Explain.

Short on time? See the most repeated AKTU Machine Learning (Open Elective) questions across all years.

Other years

More AKTU Machine Learning papers on LearnAIFast.

This paper is reproduced for study purposes only. Copyright belongs to AKTU. Source: the question paper bank of ABES Institute of Technology, an AKTU-affiliated college. If you own the copyright and want it removed, email learnwithskillnest@gmail.com and we will take it down.