Seat No.: ________ Enrolment No.

___________

GUJARAT TECHNOLOGICAL UNIVERSITY

BE - SEMESTER–VII (NEW) EXAMINATION – SUMMER 2022
Subject Code:3171114 Date:08/06/2022
Subject Name:Introduction of Machine learning
Time:02:30 PM TO 05:00 PM Total Marks: 70
Instructions:
1. Attempt all questions.
2. Make suitable assumptions wherever necessary.
3. Figures to the right indicate full marks.
4. Simple and non-programmable scientific calculators are allowed.

MARKS
Q.1 (a) What do you mean by Hypothesis? 03
(b) Differentiate between supervised and unsupervised Learning 04
(c) With Suitable Example Explain techniques for dimensionality 07
Learning
reduction
Learning
Q.2 (a) What is an Ensemble Learning 03
(b) Compare Decision Tree and Linear Regression 04
(c) With Suitable Example Explain techniques for dimensionality 07
reduction
State main difference between Bays Classifier and Naïve Bays
Classifier Also Design a predictor for given Data of whether-
enjoy sport –or not using Naïve Bays Classifier.

OR
(c) Explain with Suitable Pseudo code implementation of logistic 07
regression

Q.3 (a) How bagging is used in Machine learning 03

(b) How perceptron works explain 04
(c) Explain with Suitable Pseudo code implementation of SVM. 07
OR
Q.3 (a) How Boosting is used in machine learning 03
(b) Compare ReLu and Soft Max Function used in Neural Networks. 04
(c) Explain with suitable example back propagation Algorithm. 07

Q.4 (a) Define with neat sketch Maximum Margin 03

(b) Differentiate between weights and Hyper parameter 04
(c) Give pseudo code for classification using Neural network 07
OR
Q.4 (a) What is Deep learning? 03
(b) What do you mean by Over fitting, How to Solve this Problem 04
(c) Derive bound on number of training samples require for given 07
accuracy, Define VC Dimension and Explain how to find its value
using suitable example.
1
Q.5 (a) Define Cost Function of Neural network 03
(b) Illustrate Bias and Variance problem of machine learning 04
(c) How random forest algorithm can be build using decision tree? 07
OR
Q.5 (a) Explain about Hierarchical Clustering. 03
(b) Differentiate between linear regression and logistic regression. 04
(c) Explain in detail working of K-mean clustering algorithm. 07

*************

2
Seat No.: ________ Enrolment No.___________

GUJARAT TECHNOLOGICAL UNIVERSITY

BE – SEMESTER- VII EXAMINATION-SUMMER 2023
Subject Code: 3171114 Date: 26/06/2023
Subject Name: Introduction of Machine learning
Time: 10:30 AM TO 01:00 PM Total Marks: 70
Instructions:
1. Attempt all questions.
2. Make suitable assumptions wherever necessary.
3. Figures to the right indicate full marks.
4. Simple and non-programmable scientific calculators are allowed.

MARKS
Q.1 (a) What do you mean by weights? 03
(b) Compare Re. Lu. and Soft Max activation functions. 04
(c) How learning is achieved using supervised learning algorithm. 07
?explain it using appropriate block diagram
Q.2 (a) What do you mean by deep learning? 03
(b) How decision tree algorithm can be used for classification 04
(c) How to apply K nearest neighbor algorithm 07
OR
(c) How Dimensionality reduction is achieved by PCA. 07

Q.3 (a) Design linear repressor using pseudo code / python code. 03
(b) Design logistic repressor using pseudo code / python code 04
(c) Using Naïve bays algorithm predict probability of golf play given 07
whether today = (Sunny, Hot, Normal, False)

OR
Q.3 (a) Why over fitting is not good 03
(b) What modification is done by naïve in bays classifier? Why? 04
(c) Design SVM classifier using pseudo code / python code 07

Q.4 (a) What do you mean by Hyper parameters? 03

(b) How feed forward neural network works Explain? 04
(c) How error reduction achieved in neural network using back 07
propagation prove it using necessary equation
OR
1
Q.4 (a) What do you mean by Neuron? 03
(b) How to apply hierarchical clustering algorithm. 04
(c) How many training Examples are needed explain using VC 07
Bound?

Q.5 (a) What is Ensemble learning 03

(b) bagging and Boosting improve decision. justify 04
(c) How to apply random forest algorithm Explain in detail 07
OR
Q.5 (a) What is feature selection? 03
(b) How feature extraction technique used for Machine learning? 04
(c) Design perceptron neural network using pseudo code / python 07
code

*************

2
 Enrolment No./Seat No_____________

GUJARAT TECHNOLOGICAL UNIVERSITY

BE - SEMESTER–VII (NEW) EXAMINATION – SUMMER 2024
Subject Code:3171114 Date:22-05-2024
Subject Name:Introduction of Machine learning
Time:02:30 PM TO 05:00 PM Total Marks:70
Instructions:
1. Attempt all questions.
2. Make suitable assumptions wherever necessary.
3. Figures to the right indicate full marks.
4. Simple and non-programmable scientific calculators are allowed.
5.
MARKS
Q.1 (a) Differentiate between Dimensionality reduction and feature selection. 03
(b) Explain the process of k-fold Cross Validation in brief. 04
(c) Explain the principle of leave-one-out Cross-Validation (LOOCV) method. 07

Q.2 (a) Briefly explain the principle of Collaborative Filtering. 03

(b) State two types of distance metrics used in kNN. 04
(c) Explain the principle of ensemble learning for decision tree. 07
OR
(c) Discuss kNN model in detail. 07

Q.3 (a) Write three types of approaches used for feature selection. 03
(b) Discuss the advantages and disadvantages of using feature selection. 04
(c) Explain Principle Component Analysis algorithm for Dimensionality 07
Reduction.
OR
Q.3 (a) Write any three key properties of SVM. 03
(b) Discuss the advantages and disadvantages of using Bayesian learning. 04
(c) Explain Naïve Bayes. 07

Q.4 (a) State two types of kernel functions used in nonlinear SVM. 03
(b) How can the use of soft margins in SVM can help to manage the presence 04
of outliers and noise in data?
(c) Explain the key components of a neural network. 07
OR
Q.4 (a) Write applications of Support Vector Machine (SVM) classifier. 03
(b) Explain the concept of margin in SVM. 04
(c) Explain the principle of backpropagation algorithm. 07

Q.5 (a) State two types of computational learning models. 03

(b) Explain the difference between bagging and boosting. 04
(c) Give at least six differences between clustering and classification. 07
OR
Q.5 (a) Briefly explain the concept of Ensemble Learning. 03
(b) Discuss advantages of using the VC dimension to measure the complexity 04
of hypothesis space.
(c) Explain the principle of agglomerative hierarchical clustering. 07

*************
Seat No.: ________ Enrolment No.___________

GUJARAT TECHNOLOGICAL UNIVERSITY

BE - SEMESTER–VII (NEW) EXAMINATION – WINTER 2021
Subject Code:3171114 Date:15/12/2021
Subject Name:Introduction of Machine learning
Time:10:30 AM TO 01:00 PM Total Marks: 70
Instructions:
1. Attempt all questions.
2. Make suitable assumptions wherever necessary.
3. Figures to the right indicate full marks.
4. Simple and non-programmable scientific calculators are allowed.
MARKS
Q.1 (a) What is Machine learning? how do machines learn? 03
(b) Enlist different applications of machine learning. 04
(c) What is supervised learning? Why it is called so? 07

Q.2 (a) Explain Overfitting in context of machine learning models. 03

(b) Explain,in details, the process of K-fold cross-validation. 04
(c) Explain qualitative & quantitative data in details. Differentiate between 07
the two.
OR
(c) What is feature selection? Why is it needed? What are the different 07
approaches of feature selection?
Q.3 (a) Differentiate between Dimensionality reduction & Feature selection. 03
(b) Explain the process of evaluating a linear regression model. 04
(c) Explain, with an example, the main underlying concept of feature 07
extraction.What are the most popular algorithms for feature
extraction?
OR
Q.3 (a) What is Naïve Bayes classifier? Why is it named so? 03
(b) Enlist limitations of Bayes classifier. 04
(c) Discuss the kNN model in detail. 07
Q.4 (a) Write any two features of Bayesian learning methods. 03
(b) Explain, in brief, the decision tree algorithm 04
(c) Explain the concept of Prior and Posterior with suitable example. 07
OR
Q.4 (a) What is bootstrap aggregation? 03
(b) Explain, in brief, the SVM model. 04
(c) Explain logistic regression in detail 07

Q.5 (a) Define kernel in the SVM model. 03

(b) What is the function of a summation junction of a neuron? What is 04
threshold activation function?
(c) Write a short note on K-means clustering. 07
OR
Q.5 (a) List down various types of Ensemble learning algorithms. 03
(b) What are the advantages and disadvantages of neural networks? 04
(c) Why a multilayer neural network is required? What are the steps in 07
backpropagation algorithm?
*************

1
Seat No.: ________ Enrolment No.___________

GUJARAT TECHNOLOGICAL UNIVERSITY

BE - SEMESTER–VII (NEW) EXAMINATION – WINTER 2023
Subject Code:3171114 Date:08-12-2023
Subject Name: Introduction of Machine learning
Time: 10:30 AM TO 01:00 PM Total Marks:70
Instructions:
1. Attempt all questions.
2. Make suitable assumptions wherever necessary.
3. Figures to the right indicate full marks.
4. Simple and non-programmable scientific calculators are allowed.

MARKS

Q.1 (a) Give comparison between Supervised Learning and Unsupervised 03

Learning.
(b) Give reasons for overfitting and provide options to resolve overfitting 04
problem.
(c) Discuss K- Nearest Neighborhood classification method with its advantages 07
and disadvantages.

Q.2 (a) Write down names of three approaches for feature selection. 03
(b) At a certain University,4% of men are above 6 feet tall and 1% of women 04
are over 6 feet tall. The total student population is divided in the ratio of 3:2
in favor of woman. If a student is selected at random from among all those
over six feet tall, what is the probability that the student is woman?
(c) Write three techniques to measure feature redundancy and explain each in 07
detail.
OR
(c) A mechanical factory production line is manufacturing bolts using three 07
machines, A, B and C. The total output, machine A is responsible for 25%,
machine B for 35% and machine C for the rest. The machines that 5% of
the output from machine A is defective, 4% from machine B and 2% from
machine C. A bolt is chosen at random from the production line and found
to be defective. What is the probability that it came from (1) Machine A (2)
Machine B (3) Machine C.
Q.3 (a) Write down applications of Support Vector Machine (SVM) classifier. 03
(b) Give advantages and disadvantages of K – means Clsutering algorithm. 04
(c) Differentiate between Clustering and Classification. 07
OR
Q.3 (a) Write down desirable properties of Clustering Algorithm. 03
(b) Give a comparison between SVM and Neural Network. 04
(c) Explain the need for Kernel Method in SVM and explain use of Kernel 07
Method in SVM.
Q.4 (a) Write advantages of Neural Network. 03
(b) Write down the steps for the selection of number of hidden units for 04
backpropagation method.
(c) Explain Adaptive Linear Neuron (ADALINE) Network Model in detail. 07
OR
Q.4 (a) Write application of Neural Network. 03
(b) Write the advantages and disadvantages of Backpropagation method. 04
(c) Explain Multi Layer Feed Forward Network in detail. 07
Q.5 (a) Define Cross Validation. 03
(b) Explain Variance reduction and Bias reduction in the context of Ensemble 04
1
 Method.
(c) Explain Vapnik- Chervonenkis (VC) dimension in detail. 07
OR
Q.5 (a) Write equations for Accuracy rate, recall, and Specificity in terms of True 03
Positive (TP), true Negative (TN), False positive (FP), and False Negative
(FN).
(b) Explain Voting and Stacking approaches for Combining Ensemble 04
methods.
(c) Write down the steps for Bagging and mention its advantages and 07
disadvantages.

*************