COMPUTER SCIENCE &ENGINEERING

DEPARTMENT OF
TECHNOLOGY,. GITAM (Deemed to be Universit;
GITAM INSTITUTE OF VISAKI, HA=ATNAM-45
M.Tech.1"Sem. 2nd MID (2023-24)
Max. Marks: 30
Subject: Machine Learning Duration : 60 min
Date : 27.09.2023

Unit-I 10marks
sofclassifier building using k-Nn with the help of an
KExplain the process
example. (or)
Decision Tree Induction
2. Discuss the concept of Classifier generation through
with an example.
Unit-II 10 marks
class Classification with suitable examples.
3. Discuss the concept of Multi (or)
Regression. Compare and Contrast with Linear Regressior.
A Explain Logistic
been Unit-II 10 marks.
methodsto do so.
5. Discuss need for Regularisation in Regression and various
(or)
6 Discuss evaluation metrics for Regression iodels and comare
them.
 DepartmentGITAM
of Computer
Science and
GITAM School TechnologyEngineering
of
Course (Deemed to be University)
M.Tech. (CSE/DS/CFIS)
Subject
Machine Learing Date
Duration
06.11.2023
90 Mins.
Max Marks :30
All Questions are
mandatory and carry equal marks
1a) Write about the classification of different
clustering algorithms. (4M)
b) Apply K- Means clustering for the following
example (3 iterations). (6M)
{(2,10), (2,5), (8,4), (5,8), ( 7,5), (6,4), (1,2),(4,9) }
Assume initial clusters to the (2,10), (5,8) and (1,2)
Specify the distance metric used by you
(OR)
2. (a) List out the disadvantages of K Means algorithm (4M)

(b) Analyze how hierarchicalclustering wÙrks better than K. Means algorithm

through word's algorithm (6M)

3. (a)Analyze in details the applicatipns that fequire dimensionality reduction (4M)

At) Explainthe PCA dimensionality reduction algorithm in detail. (6M)

(OR)

4. (a) What are the situations wher¢ kernel PCA is preferred over linear PCA? Justify
(4M)
(b)Discuss kernel PCA algorithm in detail (6M)
clustering
5/Explain the concept of similarity measures in (10M)
(OR)

overfitting in clustering.
6. Explain the situation leading io (10M)
 Department of Computer Science and
GITAM School of Technology
GITAM (Deemed to be University)
Engineering
M.Tech. ( CSE/DS/CFIS) Date
Course
Subject
Machine hearning Duration 29.11.2023
50 Mins.
Max Marks: 30

Answer allQuestions. AllQuestions carTy equal marks.

Elucidate the significance of ensemble learners.
(SM)
bYCompare and contrast Bagging with Boosting
(SM)
(OR)
2. (a) Expla1n with a clear example working of RandomForests.
(SM)
(b) Discuss how random forests result in scaBable classification models
(SM)

(OR)
3. Write a Python program to handle missing values and categorical data (10M)

(OR)
4. Discuss k-feld cross validation present why it is a (10M)
viable model.
Discuss the significance of semi-supervised learning and various methods to
implement them. (10M)

(OR)
6 Discuss with1a program vith a program about hyer parameter timing. (10M)

i5