[go: up one dir, main page]
Scribd
Upload
80 views3 pages

Machine Learning Exam Paper 2024-25

This document is a Machine Learning question paper for B.Tech Semester VII at Parul University, designed for the Summer/Winter 2024-25 examination. It includes two sections with compulsory questions, covering various topics aligned with Bloom's Taxonomy levels from Remembering to Creating. The paper assesses knowledge on concepts like learning paradigms, algorithms, model evaluation, and clustering techniques.

Uploaded by

springflowers0606
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
80 views3 pages

Machine Learning Exam Paper 2024-25

This document is a Machine Learning question paper for B.Tech Semester VII at Parul University, designed for the Summer/Winter 2024-25 examination. It includes two sections with compulsory questions, covering various topics aligned with Bloom's Taxonomy levels from Remembering to Creating. The paper assesses knowledge on concepts like learning paradigms, algorithms, model evaluation, and clustering techniques.

Uploaded by

springflowers0606
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Absolutely!

Here's a third Machine Learning question paper based on your syllabus and previous
paper analysis, with Bloom’s Taxonomy (BT) levels carefully applied — from Remembering to
Creating.

📝 PARUL UNIVERSITY

FACULTY OF ENGINEERING & TECHNOLOGY


[Link] – Semester VII
Summer/Winter 2024–25 Examination
Subject Code: 303105353
Subject Name: Machine Learning
Time: 2 Hours 30 Minutes
Total Marks: 60

🔹 Instructions:

1. This question paper comprises two sections (A & B). Write answers of both sections in
separate answer books.

2. From Section A, Q.1 is compulsory. From Section B, Q.1 is compulsory.

3. Figures to the right indicate full marks.

4. Draw neat diagrams where necessary.

5. Bloom's Taxonomy Levels:

o Remember – 1

o Understand – 2

o Apply – 3

o Analyze – 4

o Evaluate – 5

o Create – 6

✅ SECTION - A

Q.1 (Compulsory – All three sub-questions) [6 + 6 = 12]

A. Define the following (BT-1):


(i) Learning Paradigm
(ii) Noise in data
(iii) Outlier

B. List and differentiate (BT-2 & BT-4):


(i) PAC Learning vs Concept Learning
(ii) Data Cleaning vs Data Transformation
(iii) General Hypothesis vs Specific Hypothesis

Q.2

A. Explain the steps involved in the Candidate Elimination Algorithm. Draw the version space
boundaries. (BT-2) [04]
B. Apply Find-S algorithm to a dataset of your choice and show step-by-step hypothesis formation.
(BT-3) [05]

OR

B. Analyze the impact of noisy and incomplete data on supervised learning performance. (BT-4)
[05]

Q.3

A. Identify and justify key steps in preprocessing a real-world dataset (e.g., student marks, health
data). (BT-3) [04]
B. Evaluate the role of normalization and feature scaling in ML model training. Use examples. (BT-
5) [05]

OR

B. Create a flowchart for a Machine Learning pipeline starting from data gathering to evaluation.
(BT-6) [05]

✅ SECTION - B

Q.1 (Compulsory – All three sub-questions) [6 + 6 = 12]

A. Define the following (BT-1):


(i) Support Vector Machine
(ii) Logistic Regression
(iii) KNN

B. Compare the following (BT-2 & BT-4):


(i) Soft Margin vs Hard Margin SVM
(ii) Linear Regression vs Logistic Regression
(iii) Single-layer vs Multi-layer Perceptron

Q.2

A. Explain the working of KNN with the help of an example dataset. (BT-2) [04]
B. Analyze the decision boundary of Logistic Regression and discuss how threshold affects
classification. (BT-4) [05]

OR
B. Evaluate the suitability of Perceptron for non-linearly separable data. What alternatives would
you suggest? (BT-5) [05]

Q.3

A. Discuss the difference between various clustering techniques like K-Means, Hierarchical, and
DBSCAN. (BT-2) [04]
B. Design a simple unsupervised learning workflow for customer segmentation using clustering.
(BT-6) [05]

OR

B. Apply cross-validation and confusion matrix techniques to evaluate any classification model.
(BT-3) [05]

✅ Let me know if you'd like:

 The solutions for this paper

 A PDF/Word format

 Or another version with specific chapters emphasized

I’ve got you covered!

You might also like