Register

Number

SRM Institute of Science and Technology

Set -
College of Engineering and Technology
School of Computing
SRM Nagar, Kattankulathur – 603203, Chengalpattu District, Tamil Nadu
Academic Year: 2024-25 (EVEN)
Test: FT4 Date: 29-04-2025
Course Code & Title: 21CSS303T-Data Science Duration: Two periods
Year& Sem: III Year /VI Sem Max.Marks:50

Course Articulation Matrix:

Course P P P P P P P P P PO PO PO
Outcome O O O O O O O O O 10 11 12
1 2 3 4 5 6 7 8 9
CO3 - - - - 1 - - - - - - -
CO4 - - - - 1 - - - - - - -
CO5 - - - - 1 - - - - - - -
Note: CO3 – To identify data manipulation and cleaning techniques using pandas
CO4 – To constructs the Graphs and plots to represent the data using python packages
CO5 – To apply the principles of the data science techniques to predict and forecast the outcome of real-
world problem
Part – A (10 x 1 = 10 Marks)
Instructions:
1) Answer ALL questions.
2) The duration for answering Part A is 15 minutes (this sheet will be collected after 15 minutes).
3) Encircle the correct answer.

S.N Question Mark B C P PI

s L O O Co
o
de
1 What is a recommended technique for handling datasets that do not 1 1 3 5
fit into memory?
A. Load the entire data into a list
B. Use streaming or chunking techniques
C. Increase screen resolution
D. Use nested loops
2 What parameter allows merge() to join datasets using an index 1 1 3 5
instead of a column?
A. on_index=True
B. use_index=True
C. left_index=True/right_index=True
D. by_index=True
3 What is the default method of dropna() in pandas? 1 1 3 5
A. Drops rows with missing values
B. Replaces missing values with 0
C. Drops columns with duplicates
D. Sorts data
4 What is binning in data preprocessing? 1 2 3 5
A. Filling missing values
B. Converting continuous variables into categorical bins
C. Merging two datasets
D. Sorting data by time
5 Which of the following techniques can be used to detect outliers or 1 2 3 5
noise in a dataset?
A. Pivoting
B. One-hot encoding
C. Z-score or IQR methods
 D. Data splitting
6 Which command is used to create subplots in Matplotlib? 1 1 4 5
A. plt.subplots()
B. plt.sub()
C. plt.mplot()
D. plt.subplotview()
7 What is Seaborn primarily used for? 1 1 4 5
A. Connecting APIs
B. Creating responsive websites
C. Creating statistical graphics on top of Matplotlib
D. Managing databases
8 In Seaborn, which function is used to plot pairwise relationships in a 1 1 4 5
dataset?
A. sns.relations()
B. sns.matrixplot()
C. sns.pairplot()
D. sns.gridplot()
9 What function is used to create a scatter plot in Matplotlib? 1 2 5 5
A. plt.point()
B. plt.scatter()
C. plt.dot()
D. plt.circles()
10 What is the purpose of a histogram? 1 2 5 5
A. To show relationship between two variables
B. To display data distribution and frequency
C. To visualize classification performance
D. To plot trends over time Regist
er
Numb
er
SRM Institute of Science and Technology
Set -
College of Engineering and Technology
School of Computing
 SRM Nagar, Kattankulathur – 603203, Chengalpattu District, Tamil Nadu
Academic Year: 2024-25 (EVEN SEM)
Test: FT4 Date:29-04-2025
Course Code & Title: 21CSS303T-Data Science Duration: Two periods
Year& Sem: III Year /VI Sem Max.Marks:50

Part – B (4 x 5 = 20 Marks)
Instructions: Answer ANY FOUR
Questions
Q Question Mark B C PO PI
. s L O Code
N
o
11 Explain the difference between reshaping, pivoting, and concatenating 5 2 3 5
datasets using pandas.
Ans:
 Reshaping: Changing the structure of data (e.g., melt()
converts wide to long format).
 Pivoting: Converting long data into a wide format (e.g.,
pivot() makes a column's values into new columns).
 Concatenating: Combining multiple datasets along rows or
columns (e.g., concat()).

12 Apply binning and standardization to a numerical dataset. Why are 5 3 3 5

these processes important in data preparation?
Ans:
Binning and standardization are important data preprocessing
techniques to improve the performance of machine learning models.
1. Binning: Converts continuous variables into discrete
categories to reduce noise and make patterns clearer.
o Example:
import pandas as pd
data = pd.Series([1, 5, 7, 9, 10, 14, 20])
bins = [0, 5, 10, 20]
labels = ['Low', 'Medium', 'High']
binned_data = pd.cut(data, bins=bins, labels=labels)
2. Standardization: Scales data to have a mean of 0 and
standard deviation of 1, which helps models converge faster.
o Example:
from sklearn.preprocessing import StandardScaler
scaler = StandardScaler()
standardized_data = scaler.fit_transform(data.values.reshape(-1, 1))
Why important?
 Binning: Simplifies complex data, making it easier for models
to detect patterns.
 Standardization: Ensures that all features are on the same
scale, preventing some features from dominating others in
models.

13 Compare and contrast the methods of handling missing data. When 5 2 3 5

would you use each?
Ans:
Removing Missing Data:
 Method: Drop rows or columns with missing values
(dropna()).
 Use: When missing data is small and won't significantly affect
the analysis or when data loss is acceptable.
Imputation:
 Method: Fill missing values with a constant (e.g., 0), mean,
median, mode, or predicted values.
 Use: When missing data is significant and removing it would
 lead to loss of important information.
Forward/Backward Fill:
 Method: Fill missing values with the previous (or next)
available data (ffill(), bfill()).
 Use: When data is time-series or ordered, and filling missing
values with neighboring data is logical.
Predictive Imputation (e.g., using ML):
 Method: Use machine learning algorithms to predict missing
values based on other features.
 Use: When missing data is substantial and imputation needs to
be more sophisticated.

14 Demonstrate how to generate a 3D surface plot using Matplotlib. 5 3 4 5

Mention the required imports and customization options.
Ans:
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D

# Create data
X = np.linspace(-5, 5, 100)
Y = np.linspace(-5, 5, 100)
X, Y = np.meshgrid(X, Y)
Z = np.sin(np.sqrt(X**2 + Y**2))

# Create a figure and 3D axis

fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')

# Plot the surface

ax.plot_surface(X, Y, Z, cmap='viridis')

# Customize labels
ax.set_xlabel('X axis')
ax.set_ylabel('Y axis')
ax.set_zlabel('Z axis')

# Show plot
plt.show()
Customization Options:
cmap: Color map for the surface (e.g., 'viridis', 'plasma').

ax.set_xlabel(), ax.set_ylabel(), ax.set_zlabel(): Customize axis labels.

ax.plot_surface(): You can add more options like edgecolor, alpha for
transparency, etc.
15 Use Seaborn to create a pairplot and customize its style using 5 3 5 5
sns.set_style() on iris dataset. What insights can a pairplot provide?
Ans:
import seaborn as sns
import matplotlib.pyplot as plt

# Load the Iris dataset

iris = sns.load_dataset('iris')

# Set the style for the plot

sns.set_style('whitegrid')

# Create a pairplot
sns.pairplot(iris, hue='species')

# Show the plot

 plt.show()
Customization:
sns.set_style('whitegrid'): Sets the plot background to white with a
grid, which enhances readability.

hue='species': Colors the points according to the different species of

the Iris flower, which helps in visualizing the relationship between
features across categories.

Insights Provided by a Pairplot:

Relationships between Variables: Shows scatter plots between each
pair of features (e.g., Sepal Length vs. Sepal Width), allowing you to
identify correlations.

Distributions: The diagonal plots (histograms or KDEs) show the

distribution of each feature.

Cluster Patterns: Helps detect if species clusters are separable based on

the features (e.g., the species may be visually separable in certain
feature combinations).

Part – C (2 x 10 = 20 Marks)
Instructions: Answer ALL questions.

Q. Question Mark BL C P PI
No s O O Code
16 Describe and compare various techniques used to clean and prepare 10 2 3 5
a raw datasets for analysis. Include examples of handling missing
data, standardization, string cleaning, and binning. Give python
code examples of each.

Ans: 1. Handling Missing Data

 Method: Removing or imputing missing values.
 Example:
o Remove rows with missing data:

import pandas as pd
df = pd.DataFrame({'A': [1, 2, None, 4], 'B': [5, None, 7, 8]})
df_cleaned = df.dropna() # Remove rows with any missing values
o Impute missing data:

df_imputed = df.fillna(df.mean()) # Replace missing with column

mean
2. Standardization (Scaling)
 Method: Scale features to have a mean of 0 and a standard
deviation of 1.
 Example:
from sklearn.preprocessing import StandardScaler
scaler = StandardScaler()
df_scaled = scaler.fit_transform(df[['A', 'B']])
3. String Cleaning
 Method: Remove or replace unwanted characters,
whitespace, or patterns from string columns.
 Example:
df['Name'] = df['Name'].str.strip().str.replace(r'\d+', '') # Remove
digits and whitespace
4. Binning (Discretization)
 Method: Convert continuous variables into categorical
bins.
 Example:
 df['Age'] = pd.cut(df['Age'], bins=[0, 18, 35, 50, 100],
labels=['Child', 'Young', 'Adult', 'Senior'])

(OR)

16 Write and explain a complete data transformation workflow using a 10 3 3 5

b sample dataset that includes missing values, text inconsistencies,
numeric scaling, and outliers. Give examples using python code.
1. Ans: Load the Dataset:

import pandas as pd
import numpy as np

# Sample data with missing values, text inconsistencies, and outliers

data = {
'Age': [25, np.nan, 22, 35, 110, 29, 200],
'Salary': [50000, 60000, np.nan, 45000, 120000, 70000, 400000],
'Name': ['John Doe', ' Jane smith ', 'alice johnson', 'BOB', 'alice',
' john', ' jane'],
'City': ['New York', 'Los Angeles', 'New York', np.nan, 'San
Francisco', 'New York', 'Miami']
}
df = pd.DataFrame(data)
1. Handle Missing Values:
 Impute missing values with appropriate methods (mean for
numeric, mode for categorical).

# Impute missing numeric values

df['Age'] = df['Age'].fillna(df['Age'].mean())
df['Salary'] = df['Salary'].fillna(df['Salary'].median())

# Impute missing categorical values

df['City'] = df['City'].fillna(df['City'].mode()[0])
2. Text Cleaning:
 Standardize text data by removing extra spaces, converting
to lowercase, etc.

# Clean and standardize text data

df['Name'] = df['Name'].str.strip().str.title() # Capitalize names and
remove leading/trailing spaces
df['City'] = df['City'].str.strip().str.title() # Ensure consistent city
names
3. Handle Outliers:
 Identify and remove outliers using the IQR method.

# Identifying outliers in 'Age' and 'Salary' using IQR

Q1_age = df['Age'].quantile(0.25)
Q3_age = df['Age'].quantile(0.75)
IQR_age = Q3_age - Q1_age
lower_bound_age = Q1_age - 1.5 * IQR_age
upper_bound_age = Q3_age + 1.5 * IQR_age

# Remove outliers
df = df[(df['Age'] >= lower_bound_age) & (df['Age'] <=
upper_bound_age)]
4. Numeric Scaling (Standardization):
 Standardize numeric columns like 'Age' and 'Salary' to
have zero mean and unit variance.
 from sklearn.preprocessing import StandardScaler

scaler = StandardScaler()
df[['Age', 'Salary']] = scaler.fit_transform(df[['Age', 'Salary']])
Final Dataframe:

print(df)

17 Explain how Matplotlib helps in customizing plots. Describe how to 10 2 4 5

a control axes, add labels, legends, annotations, and apply plot styles
with examples.
Explain the differences and use-cases of different plot types: Line
plot, Bar chart, Histogram, Box plot, Scatter plot, and Pair plot.
Ans:

Matplotlib provides powerful customization options for

creating and enhancing plots. You can control various elements
like axes, labels, legends, annotations, and styles. Here's how to
customize these features:
1. Controlling Axes:
 You can control the axis limits, ticks, and labels using
set_xlim(), set_ylim(), and set_xticks()/set_yticks().

import matplotlib.pyplot as plt

x = [1, 2, 3, 4]
y = [1, 4, 9, 16]

plt.plot(x, y)
plt.xlim(0, 5) # Set x-axis limit
plt.ylim(0, 20) # Set y-axis limit
plt.show()
2. Adding Labels and Title:
 xlabel(), ylabel(), and title() are used to add labels and
titles.

plt.plot(x, y)
plt.xlabel('X-axis Label')
plt.ylabel('Y-axis Label')
plt.title('Plot Title')
plt.show()
3. Legends:
 Use legend() to add a legend to the plot. You can label
your plots during plotting and then call legend().

plt.plot(x, y, label='y = x^2')

plt.legend()
plt.show()
4. Annotations:
 Use annotate() to add text or markers to specific points
on the plot.

plt.plot(x, y)
plt.annotate('Peak', xy=(2, 4), xytext=(3, 5),
arrowprops=dict(facecolor='red', arrowstyle="->"))
plt.show()
5. Applying Plot Styles:
 Use plt.style.use() to apply predefined styles such as
ggplot, seaborn, etc.
 plt.style.use('ggplot')
plt.plot(x, y)
plt.show()

Different Plot Types and Their Use-Cases

1. Line Plot:
o Use-case: Ideal for showing trends over time or
continuous data.
o Example: Plotting stock prices or temperature
changes.

plt.plot([1, 2, 3, 4], [1, 4, 9, 16])

plt.show()
2. Bar Chart:
o Use-case: Useful for comparing quantities across
different categories (categorical data).
o Example: Comparing sales across different
products.

plt.bar(['A', 'B', 'C'], [3, 7, 2])

plt.show()
3. Histogram:
o Use-case: Shows the distribution of data, often
for continuous numerical data.
o Example: Displaying the distribution of ages in a
dataset.

plt.hist([1, 2, 2, 3, 3, 3, 4], bins=4)

plt.show()
4. Box Plot:
o Use-case: Useful for visualizing the distribution
of data, including outliers, median, and
quartiles.
o Example: Analyzing the spread of test scores.

plt.boxplot([1, 2, 3, 4, 5, 6, 7])
plt.show()
5. Scatter Plot:
o Use-case: Displays relationships between two
variables, useful for correlation analysis.
o Example: Visualizing the relationship between
height and weight.

plt.scatter([1, 2, 3, 4], [1, 4, 9, 16])

plt.show()
6. Pair Plot:
o Use-case: Used for visualizing relationships
between multiple variables in a dataset.
o Example: Showing pairwise relationships in the
Iris dataset.

import seaborn as sns

iris = sns.load_dataset('iris')
sns.pairplot(iris, hue='species')
plt.show()

(OR)
17 Apply advanced Seaborn visualizations to explore patterns in a real 10 3 5 5
b dataset. Include pair plots, heatmaps, and style settings. Write a
Python program to visualize a 3D surface plot. Explain each
component used in the plot.
Ans:

import seaborn as sns

import matplotlib.pyplot as plt

# Load dataset
iris = sns.load_dataset('iris')

# Set style
sns.set_style("whitegrid")

# Pair plot
sns.pairplot(iris, hue='species')
plt.show()

# Heatmap (correlation matrix)

corr = iris.drop('species', axis=1).corr()
sns.heatmap(corr, annot=True, cmap='coolwarm')
plt.title('Feature Correlation Heatmap')
plt.show()

Explanation:
sns.set_style(): Sets plot background style.

pairplot(): Shows pairwise relationships and class separation.

heatmap(): Highlights correlations between numeric features.

3D Surface Plot with Matplotlib

import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D

# Data for the surface

X = np.linspace(-5, 5, 100)
Y = np.linspace(-5, 5, 100)
X, Y = np.meshgrid(X, Y)
Z = np.sin(np.sqrt(X**2 + Y**2))

# Create 3D plot
fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')

# Plot surface
surf = ax.plot_surface(X, Y, Z, cmap='viridis')

# Add labels
ax.set_xlabel('X Axis')
ax.set_ylabel('Y Axis')
ax.set_zlabel('Z Axis')
plt.title('3D Surface Plot')
plt.show()

Explanation:
Axes3D: Enables 3D plotting.

meshgrid: Generates grid for surface.

plot_surface: Draws the 3D surface.

cmap: Applies color styling to surface.

Course Outcome (CO) and Bloom’s level (BL) Coverage in Questions:

CO Coverage
60 53 %
50
40
30 26 %
21 %
20
10
0
CO 1 CO 2 CO 3