CODE COURSE NAME CATEGORY L T P CREDIT

DATA SCIENCE &

20MCA201 CORE 3 1 0 4
MACHINE LEARNING

Preamble: This is an introductory course on data science and basic concepts behind various
machine learning techniques. Machine learning is the study of adaptive computational systems
that improve their performance with experience. At the end of the course the students should
be able to design and implement machine learning solutions to classification, regression, and
clustering problems and to evaluate and interpret the results of the algorithms.

Prerequisite: Probability and Statistics, Linear Algebra, Programming in Python/R.

Course Outcomes: After the completion of the course the student will be able to:

Bloom's Category
CO No. Course Outcome (CO)
Level

Level 2:
Discuss the fundamental concepts of data science and data
CO 1
visualization techniques. Understand

Explain the basics of machine learning and use lazy learning Level 3:
CO 2 and probabilistic learning algorithms to solve data science
Apply
problems.
Describe decision trees, classification rules & regression Level 3:
CO 3 methods and how these algorithms can be applied to solve data
Apply
science problems.
Level 3:
Solve data science problems using neural networks and
CO 4
support vector machines. Apply

Discuss clustering using k-means algorithm and evaluate & Level 3:

CO 5 improve the performance of machine learning classification
Apply
models.

Mapping of Course Outcomes with Program Outcomes

PO PO PO PO PO PO PO PO PO PO PO PO
1 2 3 4 5 6 7 8 9 10 11 12
CO 1 1 1 - - - - 3 - - - - -
CO 2 3 3 3 2 - - 3 - - - - -
CO 3 3 3 3 2 - - 3 - - - - -
CO 4 3 3 3 2 - - 3 - - - - -
CO 5 3 3 3 2 - - 3 - - - - -
3/2/1: High/Medium/Low

2
Assessment Pattern

Continuous Assessment
Tests End Semester
Bloom’s Category
Examination
1 2
Remember (K1) 15 10 10
Understand (K2) 25 20 30
Apply (K3) 10 20 20
Analyse (K4)
Evaluate (K5)
Create (K6)

Mark Distribution

Total Marks CIE ESE ESE Duration

100 40 60 3 hours

Continuous Internal Evaluation Pattern:

Attendance : 8 marks
Continuous Assessment Test (2 numbers) : 20 marks
Assignment/Quiz/Course project : 12 marks

End Semester Examination Pattern: There will be two parts: Part A and Part B. Part A
contains 10 compulsory short answer questions, 2 from each module. Each question carries 3
marks. Part B contains 2 questions from each module, of which the student should answer any
one. Each question can have a maximum of 2 subdivisions and carry 6 marks.

Sample Course Level Assessment Questions

Course Outcome 1 (CO1):

1. What is data science and why do we need data science?
2. Explain the data science classification and illustrate data science tasks.
3. Describe the various methods to understand data.
4. Explain the typical methods to visualize data.

3
Course Outcome 2 (CO2)
1. Explain the differences between supervised and unsupervised machine learning
algorithms.
2. Describe the key concepts that define nearest neighbour classifiers, and why they are
considered "lazy" learners.
3. Explain how to apply k-NN classifier in a data science problem.

4. State Bayes' theorem in statistics. Outline the Naive Bayes algorithm to build
classification models.

5. Use Naive Bayes algorithm to determine whether a red domestic SUV car is a
stolen car or not using the following data:

Example Colour Type Origin Stolen?

1 red sports domestic yes
2 red sports domestic no
3 red sports domestic yes
4 yellow sports domestic no
5 yellow sports imported yes
6 yellow SUV imported no
7 yellow SUV imported yes
8 yellow SUV domestic no
9 red SUV imported no
10 red sports imported yes

Course Outcome 3 (CO3):

1. Classify data science tasks using decision trees and classification rule learners.
2. Discuss the various feature selection measures.
3. How to simplify a decision tree by pruning.
4. Describe how to construct classification rules from decision trees.
5. Explain the concepts of regression and correlation.
6. How to estimate a linear regression model.
7. Consider the following set of training examples:

Instance Classification a1 a2
1 + T T
2 + T T

4
 3 - T F
4 + F F
5 - F T
6 - F T

a) Find the entropy of this collection of training examples with respect to the target
function “classification”?
b) Calculate the information gain of a2 relative to these training examples?

Course Outcome 4 (CO4):

1. Explain how artificial neural networks mimic human brain to model arbitrary
functions and how these can be applied to real-world problems.

2. Describe different activation functions and network topology.

3. Discuss basic idea behind the backpropagation algorithm.

4. Explain how a support vector machine can be used for classification of linearly
separable data.

5. How to compute the distance of a point from a hyperplane.

6. How the kernel trick is used to construct classifiers in nonlinearly separated data.

Course Outcome 5 (CO5):

1. Explain how the clustering tasks differ from the classification tasks.

2. How clustering defines a group, and how such groups are identified by k-means
clustering algorithm.

3. Find the three clusters after one epoch for the following eight examples using the k-
means algorithm and Euclidean distance: A1 = (2,10), A2 = (2,5), A3 = (8,4), A4 =
(5,8), A5 = (7,5), A6 = (6,4), A7 = (1,2), A8 = (4,9). Suppose that the initial seeds
(centres of each cluster) are A1, A4 and A7.

4. Explain the various matrices used to measure the performance of classification

algorithms

5. Explain the concepts of bagging and boosting.

6. Suppose 10000 patients get tested for flu; out of them, 9000 are actually healthy and
1000 are actually sick. For the sick people, a test was positive for 620 and negative
for 380. For the healthy people, the same test was positive for 180 and negative for
8820. Construct a confusion matrix for the data and compute the precision and recall
for the data.

5
 Model Question Paper
Course Code: 20MCA201

Course Name: DATA SCIENCE AND MACHINE LEARNING

Max. Marks :60 Duration: 3 Hrs

Part A

Answer all questions. Each question carries 3 marks (10 * 3 = 30 Marks)

1. What is data science?

2. Explain the different types of data.
3. Differentiate between supervised and unsupervised learning algorithms.
4. Explain how to choose the value of k in k-NN algorithm.
5. Explain entropy and information gain.
6. Explain the Ordinary Least Square method in regression.
7. Define activation function. Give two examples.
8. What is maximum margin hyperplane.
9. Define precision, recall and F-measure.
10. Explain bootstrap sampling

Part B

Answer one full question from each module, each carries 6 marks.

11. Explain the various methods for visualising multivariate data. (6 marks)

OR

12. Explain the various processes for preparing a dataset to perform a data science
task. (6 marks)

13. Based on a survey conducted in an institution, students are classified based on the
two attributes of academic excellence and other activities. Given the following data,
identify the classification of a student with X = 5 and Y = 7 using k-NN algorithm
(choose k as 3).

X (Academic Excellence) Y (Other Activities) Z (Classification)

8 6 Outstanding

5 6 Good

7 3 Good

6 9 Outstanding

(6 marks)

OR

6
14. Given the following data on a certain set of patients seen by a doctor. Can the doctor
conclude that a person having chills, fever, mild headache and without running nose
has flu? (Use Naive Bayes classification).

Chills Running nose Headache Fever Has flu

Y N mild Y N

Y Y no N Y

Y N strong Y Y

N Y mild Y Y

N N no N N

N Y strong Y Y

N Y strong N N

Y Y mild Y Y

(6 marks)

15. Obtain a linear regression for the data given in the table below assuming that y is
the independent variable.

x 55 60 65 70 80

y 52 54 56 58 62

(6 marks)

OR

16. Given the following data, draw a decision tree to predict whether a person cheats.
Give the corresponding set of classification rules also.

Sl. No. Refund Marital status Income Cheats?

1 Yes Single High No

2 No Married High No

3 No Single Low No

4 Yes Married High No

5 No Divorced High Yes

6 No Married Low No

7
 7 Yes Divorced High No

8 No Single High Yes

9 No Married Low No

10 No Single High Yes

(6 marks)

17. Define an artificial neuron. What are the characteristics of an artificial neural
network (ANN)?
(6 marks)
OR

18. a) Define linearly separable dataset. Give an example each of a dataset that is
linearly separable and of a dataset that is not linearly separable.
(3 marks)

b) Define kernel function. Explain the kernel trick to construct a classifier for a
dataset that is not linearly separable.
(3 marks)

19. Suppose 10000 patients get tested for flu; out of them, 9000 are actually healthy
and 1000 are actually sick. For the sick people, a test was positive for 620 and
negative for 380. For the healthy people, the same test was positive for 180 and
negative for 8820. Construct a confusion matrix for the data and compute the
precision and recall for the data. (6 marks)

OR
20. Assume the following: A database contains 80 records on a particular topic of which
55 are relevant to a certain investigation. A search was conducted on that topic and
50 records were retrieved. Of the 50 records retrieved, 40 were relevant. Construct
the confusion matrix for the search and calculate the precision and recall scores for
the search. (6 marks)

8
 Syllabus

Module 1 (9 Hours)

Introduction to data science, Data science classification, Data science process - Prior
knowledge, Data preparation, Modelling, Application, Data exploration - Data sets,
Descriptive statistics for univariate and multivariate data
Data visualisation – Histogram, Quartile plot, Distribution chart, Scatter plot, Bubble chart,
Density chart

Module 2 (9 Hours)

Introduction to machine learning: How machines learn - Data storage, Abstraction,

Generalisation, Evaluation, Machine learning in practice - Types of machine learning
algorithms.
Lazy learning: Classification using K-Nearest Neighbour algorithm - Measuring similarity
with distance, Choice of k, Preparing data for use with k-NN.
Probabilistic learning: Understanding Naive Bayes - Conditional probability and Bayes
theorem, Naive Bayes algorithm for classification, The Laplace estimator, Using numeric
features with Naive Bayes.

Module 3 (9 Hours)

Decision tree learning: Concept of decision tree, Divide and conquer approach, C5.0
Decision tree algorithm, Choosing the best split, Pruning the decision tree.
Classification rules learning: Concept of classification rules, Separate and conquer
approach, The 1R algorithm, Rules from decision trees.
Regression methods: Concept of regression, Simple linear regression, Ordinary least
squares estimation, Correlations, Multiple linear regression.

Module 4 (9 Hours)

Neural network learning: Artificial neurons, Activation functions, Network topology,

Training neural networks with backpropagation.
Support vector machines: Hyperplanes, Classification using hyperplanes, Maximum
margin hyperplanes in linearly separable data, Using kernels for non-linear spaces.

Module 5 (9 Hours)

Clustering: The k-means clustering algorithm, Using distance to assign and update clusters,
Choosing number of clusters.
Evaluating model performance: Confusion matrices, Precision and recall, Sensitivity and
specificity, Precision and recall, F-measure, ROC curves, Cross validation - K-fold cross
validation, Bootstrap sampling.
Improving model performance - Bagging, Boosting, Random forests.

9
Text Books

1. Vijay Kotu, Bala Deshpande, Data Science Concepts and Practice, Morgan Kaufmann
Publishers 2018 (Module 1)
2. Brett Lantz, Machine Learning with R, Second edition, PackT publishing 2015
(Modules 2 to 5)

Reference Books

1. Michael Steinbach, Pang-Ning Tan, and Vipin Kumar, Introduction to Data Mining,
Pearson 2016.
2. Jiawei Han, Micheline Kamber and Jian Pei, Data mining Concepts and techniques,
Morgan Kaufmann Publishers 2012
3. Peter Harrington, Machine Learning in action, Dreamtech publishers 2012
4. Dr M Gopal, Applied Machine learning, McGraw Hill Education Private Limited
5. E. Alpayidin, Introduction to Machine Learning, Prentice Hall of India (2005)
6. T. Hastie, RT Ibrashiran and J. Friedman, The Elements of Statistical Learning,
Springer 2001
7. Data Science from Scratch: First Principles with Python, Joel Grus, O’Reilly, First
edition, 2015
8. Introducing Data Science, Davy Cielen, Arno D. B. Meysman, Mohamed Ali, Manning
Publications Co., 1st edition, 2016

Web Resources:

1. https://www.coursera.org/learn/machine-learning
2. https://www.coursera.org/learn/data-scientists-tools

Course Contents and Lecture Schedule

No. of
Topic
Lectures

1 Module 1 9 hrs

1.1 Introduction to data science - What is data science? Why data science? 2 hrs

1.2 Data science classification 1 hr

Data science process - Prior knowledge, Data preparation, Modelling,

1.3 2 hrs
Application
Data exploration- Data sets, Descriptive statistics for univariate and
1.4 2 hrs
multivariate data
Data visualization – Histogram, Quartile plot, Distribution chart, Scatter
1.5 2 hrs
plot, Bubble chart, Density chart

2 Module 2 9 hrs

10
 How machines learn – Data storage – Abstraction – Generalisation –
2.1 1hr
Evaluation

2.2 Machine learning in practice – Types of machine learning algorithms. 1hr

2.3 Classification: Lazy learning - K-Nearest Neighbour algorithm 2 hrs

2.4 Measure of similarity, Choice of k 1 hr
2.5 Preparing data for use with k-NN 1 hr

2.6 Probabilistic Learning: Conditional probability and Bayes theorem. 1 hr

2.7 Naive Bayes algorithm 2 hrs

3 Module 3 9 hrs

3.1 Concept of decision tree, Divide and conquer approach 1 hr

3.2 C5.0 Decision tree algorithm 1 hr
3.3 Choosing the best split, Pruning the decision tree 2 hrs
Classification rules learning: Concept of classification rules, Separate
3.4 1 hr
and conquer approach

3.5 The 1R algorithm, Rules from decision trees 1 hr

3.6 Regression methods: Concept of regression, Correlations 1 hr

3.7 Simple linear regression, Ordinary least squares estimation 1 hr

3.8 Multiple linear regression 1 hr

4 Module 4 9 hrs
4.1 Understanding neural networks - Artificial neurons 1 hr
4.2 Activation functions, Network topology 2 hrs
4.3 Training neural networks with back propagation 1hr
Understanding Support Vector Machines, Classification with
4.4 1 hr
hyperplane
4.5 Linearly separable data, Nonlinearly separable data 1 hr
Methods to find maximum margin hyperplanes in linearly separable
4.6 1 hr
data
4.7 Using kernels for non-linear spaces 2 hrs

5 Module 5 9 hrs

5.1 Understanding Clustering - The k-means clustering algorithm 1 hr

11
 Using distance to assign and update clusters, Choosing the appropriate
5.2 1 hr
number of clusters
Evaluating model performance: Confusion matrices, Precision and
5.3 recall, Sensitivity and specificity, Precision and recall, F-measure, ROC 2 hrs
curves.
5.4 Cross validation: K-fold cross validation, Bootstrap sampling 2 hrs

5.5 Improving model performance: Bagging, Boosting 2 hrs

5.6 Random forests 1 hr

12