Roll No.
KRISHNA ENGINEERING COLLEGE, GHAZIABAD
Department of CS-AI & CS-AIML
B. Tech. (SEM: VI )
CT-1 (Even Sem-2023-24), April 2024
Subject Name: Machine Learning Techniques Subject Code: KAI601
Max. Time: 2Hrs Max. Marks:50
Faculty Name: Dr. Pratibha Singh
CO 1 To understand the need for machine learning for various problem solving
CO 2 To understand a wide variety of learning algorithms and how to evaluate models generated from data
CO 4 To design appropriate machine learning algorithms and apply the algorithms to a real world problems
Note: Attempt all sections.
SECTION – A
Attempt all questions. (2x4=8)
Q. No. Question Marks CO
a What is a “Well -posed Learning “problem? Explain with an example. 2 1
b What are the components in designing a learning system? 2 1
1.
c What is the difference between linear and logistic regression? 2 2
d What the difference between linear separable and non-linear separable problems? 2 4
SECTION – B
Attempt all questions. (6x3=18)
Q. No. Question Marks CO
a. Compare regression, classification and clustering with real life examples.
OR
2. 6 1
b. What are the types of learning? How are they different from each other?
Explain it with suitable examples.
a. What is Logistic regression? How is it used for the binary classification?
OR
3. 6 2
b. What is the Bayesian Belief Network (BBN)? Explain BBN classifier with an
example.
a. What is the gradient descent and delta rule? What are its limitations?
OR
b. What is the perceptron learning rule?
4. Execute the steps of perceptron learning algorithm for maximum two iterations 6 4
for the training samples (1,1,1) and (1,0,1) that belongs to class 1, and the
training samples (1,0,0) and (0,0,1) that belongs to class 2. Assume the learning
rate 0.1 and initial weights as 0.3, -0.1, 0.2.
SECTION – C
Attempt all questions. (8x3=24)
Q. No. Question Marks CO
a. What is the K-means clustering? Use the K-means clustering and
Euclidean distance to cluster the following 5 examples:
5. (2,10), (3,5), (6,4), (1,2), (2,3) up to maximum 2 iterations. 8 1
OR
b. Compare Artificial Neural Networks (ANN) and Bayesian networks.
a. Find the final hypothesis set using Candidate Elimination Algorithm for
the following dataset:
6. 8 2
OR
b. What is Naïve Bayes algorithm? Explain the Naïve Bayes classifier with a
working example.
a. What is the gradient descent learning algorithm? What is the difference
between stochastic gradient descent and standard gradient descent
algorithms?
7. OR 8 4
b. Derive the weight updates rule between hidden layer and output layer, and
input layer and hidden layer for multilayer feedforward backpropagation
network.
