23AD311 - FUNDAMENTALS OF DATA SCIENCE

Unit – I
Part – A Questions
Artificial Intelligence and Data Science
23AD311 - FUNDAMENTALS OF DATA SCIENCE
Third Semester
Q.N CO’s Bloom’s
Questions Level
o
1 Differentiate Big Data and Data Science CO1 K2
2 List out few areas where Data Science could be applied CO1 K1
3 Give an example of how the government organization could use the data’s CO1 K2
value
4 Define Unstructured data with an example CO1 K1
5 How can we apply data science to machine-generated data CO1 K2
6 Show how we can generate data from graph-based or network data CO1 K2
7 With an example explain streaming data CO1 K2
8 List down the steps involved in the process of data science CO2 K1
9 Identify the right business question to ask to understand a project CO2 K3
10 What do you mean by data modelling or model building CO2 K1
11 What are different ways to retrieve the required data CO2 K2
12 List few open data providers CO2 K1
13 Give an example of data entry errors CO2 K2
14 What are the different ways of combining data? CO2 K2
15 What do you mean by EDA CO2 K1
16 Define outliers CO2 K1
17 Why do we need to pre-process the data? CO2 K2
18 What are different visualization techniques used in the step of EDA CO2 K2
19 How would you apply data cleansing techniques to prepare messy sales data CO2 K3
for predictive modeling in a retail company?
20 In a healthcare analytics project, how can exploratory data analysis (EDA) CO2 K3
help identify critical patient trends?
Part – B
1 How can we distinguish between big data and data science, and what are the CO1 K2
respective advantages and applications of both in driving insights and
decisions?
2 Elaborate on the steps involved in the comprehensive process of data science, CO2 K2
focusing on data retrieval, cleaning, and exploratory data analysis (EDA)?
3 Illustrate the step-by-step process involved in Exploratory Data Analysis. CO2 K2
4 Provide an in-depth exploration of the stages in the data science process, CO2 K3
emphasizing model building and visualization with a scenario.
5 Emphasize the importance of data in data science and examine different CO2 K2
facets of data and challenges in processing.
6 A telecom company is experiencing high customer churn. Build a predictive CO2 K3
model to identify influencing factors.
7 A retail chain finds discrepancies in its sales data. Improve data quality for CO2 K3
accurate analysis and decision-making.
8 A metropolitan city wants to improve its public transport system. Explore CO2 K3
how data science can handle various data types to support this.
 Unit II

Part - A
Q.N CO’s Bloom’s
Questions Level
o
1. What is NumPy? CO3 K1
2. What are the various features of NumPy? CO3 K1
3. How do you create a 3D array? CO3 K1
4. Give the code to create a 1D array and 2D array using Numpy CO3 K1
What is the procedure to count the number of times a given value appears in CO3
5. K1
an array of integers?
Create and display a numpy array containing CO3
a. zero Elements of any size array
6. K2
b. 2X5 array filled with 1’s
c. 3X7 array filled with pi’s (use np.pi)
Given the below array answer the following queries CO3
x = [[27 78 28 81 8]
[ 1 97 8 16 11]
7. [ 8 63 80 60 11]] K2

a. Reshape the above array into 5X3 size and display the array
b. Access the element at position (1,2)
Given the below array answer the following queries CO3
x = [[27 78 28 81 8]
[ 1 97 8 16 11]
[ 8 63 80 60 11]]
8. K1
a. Reshape the above array into 5X3 size and display the array
b. Access the element at position (1,2)

9. Define Universal function and list few ufuncs related arithmetic CO3 K1
Why ufuncs arithmetic operators are faster when compare to Python's native CO3
10. K1
arithmetic operators.
11. What are different types of aggregates are there in numpy CO3 K1
12. List the rules of broadcasting. CO3 K1
Is it possible to create a heterogeneous numpy array. If possible write a code CO3
13. K1
to demonstrate it.
14. What do you mean by Fancy Indexing CO3 K1
15. Define the min and max function in NumPy CO3 K1
16. How to generate random numbers with NumPy? CO3 K1
17. Define broadcasting in NumPy with an example. CO3 K2
18. What is fancy indexing in NumPy? Give a simple example. CO3 K1
19. How can you perform boolean masking in NumPy arrays? CO3 K2
 Write a NumPy command to find the maximum and minimum values of an CO3
20. K1
array.
Part – B
Illustrate the array manipulation through Attributes of arrays, Indexing of CO3
1. K2
arrays and Slicing of arrays.
Demonstrate the purpose of universal functions in NumPy with the help of
python code. List and illustrate different types of universal functions (unary
2. and binary) with examples. Compare the advantages of using universal CO3 K2
functions over traditional Python loops for array computations along with
code. The example should include comments explaining each step.
Briefly explain the NumPy’s broadcasting functionality. Give the rules of CO3
3. K2
broadcasting with examples for each of the rules.
Construct and demonstrates the use of NumPy’s structured arrays and Fancy CO3
4. K2
Indexing with examples.
Given the below array
[[5 0 3 3]
[7 9 3 5]
[2 4 7 6]]
Write a code to count the number of True entries in a Boolean array for the
following queries 8*2 = 16 Marks
a) How many values less than 6?
5. CO3 K3
b) How many values less than 6 in each row?
c) Are there any values greater than 8?
d) Are there any values less than zero?
e) Are all values less than 10?
f) Are all values equal to 6?
g) Are all values in each row less than 8?
h) Sum the values greater than 2 lesser than 5
Let’s consider the heights of all US presidents. This data is available in the
file president_heights.csv, which is a simple comma-separated list of labels
and values.
Order Name Height(Cm)
1 George Washington 189
2 John Adams 170
3 Thomas Jefferson 189
4 James Madison 163

6. The sample data is given above. K3

CO3
a. Compute a variety of summary statistics including mean, standard
deviation, min and max using Numpy libraries 4 Marks
b. Compute the Quantiles. How do you calculate the median height of
US presidents? 2 Marks
c. Plot the histogram for the heights attribute. 4 Marks
d. Write a short notes on Multidimensional Aggregates and list the other
aggregations function available 6 Marks

Explain and demonstrate the use of universal functions in NumPy with CO3
7. K3
suitable examples. Include at least five mathematical operations.
8. Write a Python program using NumPy to demonstrate the following: K3
 a) Array broadcasting CO3
b) Boolean logic and masking
c) Fancy indexing
d) Sorting structured data.
Explain the output.

Unit III

Part - A
Q.No Questions CO’s Bloom’s
Level
1 Define Pandas/Python Pandas. CO4 K1
2 Mention the different types of data structures in Pandas. CO4 K1
3 How can we calculate the standard deviation from a Series? CO4 K2
4 Define DataFrame in Pandas. CO4 K1
5 Create and display a Pandas Series using a dictionary. CO4 K3
6 What are the significant features of the Pandas library? CO4 K1
7 Create a Pandas DataFrame using a list. CO4 K3
8 Create a DataFrame from a dictionary of ndarrays. CO4 K3
9 Construct DataFrame objects using a single Series object. CO4 K3
Given the following series (data) write a code to access the element at index CO4
b and d
a 0.25
10 K3
b 0.50
c 0.75
d 1.00
11 Explain Categorical data in Pandas? CO4 K2
12 How will you create an empty DataFrame in Pandas? CO4 K3
13 What are the different types of merges available in pandas ? CO4 K1
What will be the output of the following code? CO4
A = np.arange(1, 13).reshape(3, 4)
14 K3
df = pd.DataFrame(A, columns=list('QRST'))
df
15 What happens when we add the two DataFrame as shown CO4 K3

DataFrame A DataFrame B

A 1 A 0
B 9 B 0
D 4 C 2
G 1 D 3
dtype: int64 E 2
dtype: int64
 Write Python code to select rows in a DataFrame where the column ‘Age’ > CO4
16 K3
30.
17 What is the difference between loc[] and iloc[] in Pandas? CO4 K2
18 How do you handle missing data in a Pandas DataFrame? CO4 K2
Write code to group data in a DataFrame by a column Department and CO4
19 K3
compute the average salary.
What will be the output of the following code CO4
20 vals2 = np.array([1, np.nan, 3, 4]) K3
vals2.mean()
Part – B
Illustrate in detail the various data structures available in pandas objects with CO4
1. K2
examples
Show how the NumPy array slicing, masking, and fancy indexing could be CO4
2. K2
applied on pandas series object and DataFrame object.
Demonstrate how the index alignment happens in pandas Series and perform
the following on DataFrame using ufuncs with help of python code

3. Create a DataFrame of size 4*3 with random integers CO4 K2

i. Add 10 to each element.
ii. Multiply each element by 2.
iii. Apply the cosine function to each element
Illustrate with python code on creating the Hierarchical Indexing and give the CO4
4. K2
implementation of Indexing and Slicing a MultiIndex dataset
i) Describe the various methods of handling the missing data in Pandas CO4
5. ii) Write the code showing the implementation of different types of K2
merges with examples.
List and explain the components of the time series model with a suitable CO4
6. K2
example.
Image you have a series of data that represents the amount of precipitation
each day for a year in a given city.
a) Load the daily rainfall statistics for the city of Chennai in 2024.
Which is given in a csv file Chennai rainfall 2024.csv and convert the CO4
7. K3
date column to datetime frame 5 Marks
b) Using Pandas generate a histogram for rainy days 3 Marks
c) Find out the days that have high rainfall 4 Marks
d) Generate a line plot using Seaborn 4 Marks
8. Consider that an E commerce organization like Amazon have different region K3
sales as Northsales.csv, Southsales.csv, Westsales.csv and Eastsales.csv in the
csv format. Write a code to perform the following queries
a) Load and Combine North and West region sales and Combine South
and East region sales.
4 Marks
b) Calculate the Aggregate sales for each combined region. 3 Marks CO4
c) Find the maximum and minimum sales for each combined region. 3
Marks
d) Determine the number of sales records for each combined region. 3
Marks
e) Compare the average sales between the two combined regions. 3
 Marks

Unit IV Data Visualization

Part – A

Q.No Questions CO’s Bloom’s

Level
1.
Give the command for importing matplotlib. Draw a scatterplot with green CO5 K3
dots using matplotlib
2. How to Display the Plots. Give an example to display a graph CO5 K3
3. Create a two 2-D array. Plot it using matplotlib CO5 K3
4. What do you mean by data visualization technique? CO5 K1
5. Name the library and interface used to plot a chart in python CO5 K1
6. How to plot two subplots using MATLAB-style interface CO5 K3
Write code to do the following: CO5
Plot the following data on line chart:

7. Runs in Overs 10 20 K3

MI 110 224

RCB 85 210
8. How to change the thickness of line? CO5 K2
Plot following data on bar graph: CO5

English: 56,78,90,34
9. K3
Science: 65,77,54,32

Maths: 45,67,43,41

10. Define Basic Errorbar() function in matplotlib CO5 K1

11. How to display three-dimensional data in two dimen- CO5 K2
sions
12. What is histograms and how it can called using matplotlib in python CO5 K2
13. What is Density Plot CO5 K1
14. What is the purpose of error bar function in Matplotlib? Give CO5 K2
an example.
15. Write the command for Text annotations with Example CO5 K3
16. Briefly explain Visualizing Error with example CO5 K2
17. How to save a Matplotlib figure as an image file? Provide the CO5 K3
command.
18. What is the difference between plt.figure() and plt.subplot()? CO5 K2
19. Define a contour plot and mention one use case where it is CO5 K1
 applied.
20. What is the purpose of tight_layout() in matplotlib plots? CO5 K2
Part – B
Explain about various visualization charts like line plots, CO5
1. 1Density and contour Plot, histograms using Matplotlib with an K2
example
2. 2Describe scatter plot in detail and demonstrate the scatter plot using plt.plot CO5 K2
and plt.scatter. Differentiate both the plots
3. 3Compare seaborn with matplotlib. Explore different plot types available in CO5 K4
Seaborn.
Discuss the salient features of the Bokeh library, and how it can be used for CO5
4. K2
interactive data visualization of large-scale datasets.
a. How text and image annotations are done using python? Give an example
of your own with appropriate Python code.
CO5
5. 5 K3
b. Show how three-dimensional plot can be created from sets of (x,y, z)
triples

Show the implementation code describing the below components of

matplotlib module in python
a. Create a simple figure and an axes.
b. Control the line colours and styles CO5
6. K3
c. Adjust the axis limits
d. Label the Plots
e. Highlighting the maximum and minimum points on the sine and
cosine waves.
a) Discuss the purpose and syntax of barh(), hist(), and fill_between() CO5
7. functions in matplotlib. K3
b) Provide code examples for each.
Compare Matplotlib, Seaborn, and Plotly in terms of capabilities, visual CO5
8. K4
aesthetics, and ease of use. Provide examples where each library is preferable.

Unit V Ethics and Data Science

Part – A
Q.No Questions CO’s Bloom’s
Level
1. What does it take to build a good data product or service CO6 K1
2. List the 5 framing guidelines CO6 K1
3. What do you mean by golden rule for data? CO6 K1
4. Under what situation consent could become a treat CO6 K1
5. List few major organizations which collects data without user CO6 K1
consent
6. How clarity could be related to consent. CO6 K1
7. “Consistency and trust can be broken either explicitly or CO6 K
implicitly”. Justify with example
8. Is it possible to control how the services uses our data? If so CO6 K
give examples
 List the ethical considerations for taking responsibility while CO6 K1
9. building, maintaining, and managing data, technologies, particularly in data
science, machine learning, and artificial intelligence
10. When do we need to access the social impact statement of CO6 K2
the project?
11. Relate the process “Andon cord” to data science CO6 K2
12. Suggest few methods for a Data –Driven organization to CO6 K1
escalate issues without the fear of reprisal.
13. Enumerate the Google AI principles CO6 K1
14. Mention few regulatory bodies with regard to data science in CO6 K1
India
15. How do we know that the data is valid for its intended use? CO6 K2
16. Discuss briefly on General data protection regulation CO6 K2
17. Give some relevant example why data scientist need to CO6 K2
understand data and follow data ethics
18. List few ways how an organization could build ethics into CO6 K2
culture
19. What are the Five Cs in the context of data ethics and governance? CO6 K1
20. Why is transparency important in data science practices? CO6 K2
Part – B
How can a company ensure that it obtains informed and
1. explicit consent from its customers before collecting and CO6 K2
using their data for personalized marketing strategies?
What measures can a company take to maintain transparency
2. and provide clear information to its customers about how CO6 K2
their data is collected, stored, and used?
How can a company ensure that its data collection and
3. analysis processes are applied consistently across all CO6 K2
customer interactions to maintain fairness and accuracy?
i) What according to you should be ethical principles for conducting
4. research that involves dealing with other people’s data? CO6 K2
ii) Should there be difference in expectations about what is ethical
online versus offline regarding handling of data
Consider a scenario when you check into a hospital you are
required to sign a form that gives them the right to use your CO6
5. data. You're agreeing that they can use your data in K2
unspecified ways. Defend the above statement in terms of
consent.
A retail company wants to use data science to improve the
customer shopping experience by personalizing recommend-
6. ations and optimizing inventory based on purchasing beha- CO6 K2
vior. Apply the 5 C's framework so that the retail company en-
hances the customer shopping experience while maintaining
ethical standards.
7. A fintech startup collects user financial data to offer personal- K3
ized investment advice. However, it plans to share this data
with third-party advertisers for revenue generation.
a. Discuss the ethical implications of this practice in the con- CO6
text of data ownership, user consent, and privacy.
 b. Suggest a responsible data strategy that balances business
goals and ethical data handling.
With reference to GDPR (General Data Protection Regulation)
and Indian Data Protection Laws, CO6
a. Compare the principles of data protection in both regula-
8. tions. K3
b. Explain how these laws influence the design of ethical data
science projects, especially in sectors like healthcare and
education.