Scribd
Upload
9 views10 pages

CAT2 Key

This document is a question paper for a Continuous Assessment Test in the subject of Introduction to Machine Learning. It contains questions in three parts - Part A with 6 multiple choice questions worth 2 marks each, Part B with 3 questions worth 6 marks each, and Part C with 2 questions worth 10 marks each. The questions cover topics related to machine learning algorithms like support vector machines, decision trees, ensemble methods, dimensionality reduction techniques, and their applications. The document provides the framework and questions for evaluating students' understanding of core machine learning concepts.

Uploaded by

Vikashanand Sasikumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views10 pages

CAT2 Key

This document is a question paper for a Continuous Assessment Test in the subject of Introduction to Machine Learning. It contains questions in three parts - Part A with 6 multiple choice questions worth 2 marks each, Part B with 3 questions worth 6 marks each, and Part C with 2 questions worth 10 marks each. The questions cover topics related to machine learning algorithms like support vector machines, decision trees, ensemble methods, dimensionality reduction techniques, and their applications. The document provides the framework and questions for evaluating students' understanding of core machine learning concepts.

Uploaded by

Vikashanand Sasikumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

Register Number

Sri Sivasubramaniya Nadar College of Engineering, Kalavakkam – 603 110


(An Autonomous Institution, Affiliated to Anna University, Chennai)
Department of Computer Science and Engineering
Continuous Assessment Test – II
Question Paper

Degree & Branch B.E. & CSE Semester VI


UCS1603 & INTRODUCTION TO MACHINE
Subject Code & Name Regulation: 2018
LEARNING
2022-2023
Academic Year Batch 2020-2024 Date 03.04.2023 FN
EVEN
Time: 08:15AM-9:45AM
Answer All Questions Maximum: 50 Marks
(90 Minutes)
COURSE OUTCOMES
CO1: Understand the basic concepts of machine learning (K2)
CO2: Apply supervised algorithms for different classification problems (K3)
CO3: Understand the need of ensemble methods (K2)
CO4: Apply unsupervised and reinforcement learning techniques to various problems (K3)
CO5: Understand the requirement of dimensionality reduction and optimization techniques (K2)

Part – A (6×2 = 12 Marks)


1. The optimization problem of support vector machine can be solved
using primal and dual methods. Which is preferable? Why?
Ans: Primal used when dataset is large and dimension is small,
Dual is used when dimension is large
In general dual form is preferable because most of the datasets has
huge dimensions and also programming packages support is available
for dual forms.
2. Define the polynomial and RBF kernel functions of support vector
machine.

1.4.1,
2.3.1

Ans:
3. The probabilities of five events are: P(first) = 0.5, P(second)=
P(third)=P(fourth)=P(fifth)= 0.125. Calculate entropy using the given
probabilities.
Ans:

1.4.1,
13.3.1

4. What is meant by CART? When is it preferable to use in ML model 1.4.1,


creation? 2.3.1
Ans: CART means Classification and Regression Tree.
In summary, CART is preferable when you have a large dataset with
many features, missing values, and a need to interpret the model. It's
also useful when you need to handle both categorical and continuous
variables, and when there are complex interactions between input
features.
5. Five different models are built using different features and different
samples, to find whether the given sample image is Cancerous (C) or
not Cancerous (NC). The results given by the five models are:
1.4.1,
M1→C, M2→NC, M3→C, M4 →C, M5 →NC.
2.1.3
Identify the suitable strategy to derive the final decision.
Ans: Majority Voting and the final decision is Cancerous since three
out of 5 models results are cancerous.
6. What is the need of dimensionality reduction techniques in machine
learning process?
Ans:
1.4.1,
To improve model performance,
2.1.3
To avoid the curse of dimensionality
To improve interpretability:
To speed up computation

Part – B (3×6 = 18 Marks)


7.Compare Random Forest and Boosting ensemble algorithms.

8. Construct space splitting and K-d tree representation for the following data points:
(5,4), (2,2), (3,7), (4,6), (7,5), (6,1), (6,8)

9. The pattern number and its corresponding data points (X) with label (Y) for the training set are
given in the following table.
Pattern X Y
1 1 -1
2 3 -1
3 7 +1
The SVM objective function is “g(x) = wx + b”. Using the training set, find the values of x, w
and b in the objective function. Also, validate the function, for the test data x=9 to find the class
label.
Part – C (2×10 = 20 Marks)
10. Construct the decision tree for the dataset given in the following table.
Use the impurity functions: Entropy and Information gain to draw the
first level tree representation.
Height Hair Eyes Attractive?
Small Blonde Brown No
Tall Dark Brown No
Tall Blonde Blue Yes
Tall Dark Blue No
Small Dark Blue No
Tall Red Blue Yes
Tall Blonde Brown No
Small Blonde Blue Yes
11. Construct the decision tree for the dataset given in the following table.
Use the Gini Index impurity function to draw the final decision tree.
Write the corresponding rules in the Implication form (if condition –
then).
Past Trend Open Trading Return
Interest Volume

Positive Low High Up

Negative High Low Down

Positive Low High Up

Positive High High Up

Negative Low High Down

Positive Low Low Down

Negative High High Down

Negative Low High Down

Positive Low Low Down

Positive High High Up

https://blog.quantinsti.com/gini-index/
12. Explain Linear Discriminant Analysis (LDA) as a sequence of steps
with necessary equations and state few applications of LDA.
Definition, Equations -8mark
Applications – 2 mark
13. Explain Principal Component Analysis (PCA) as a sequence of steps
with its importance in dimensionality reduction. Compare PCA with
LDA (7 + 3)
Definition, steps – 7mark
Comparison -3mark

You might also like