Indian Institute of Technology Dharwad
CS214: Artificial Intelligence Laboratory
Lab 1: Data Visualization and
Statistics from Data
Note
There are two problems that need to be worked out in the lab. Problem
Statement 1 is compulsory. Problem Statement 2 is a bonus.
Problem Statement 1
A dataset related to Indian diabetes, containing medical attributes, is pro-
vided in a CSV file (‘pima-indians-diabetes.csv‘). The dataset includes var-
ious features that provide insights into the conditions leading to diabetes.
These features are: pregs, plas, pres, skin, test, BMI, pedi, Age, and
class.
Write a Python program to perform the following tasks:
1. Display the first 10 tuples of the given dataset using the ‘head()‘ func-
tion.
2. Display the structure of the data to provide details about the number of
entries, data types, and memory usage of the dataset using the ‘info()‘
function.
3. Generate the descriptive statistics for each numerical attribute using
the ‘describe()‘ function to provide the count, mean, standard devia-
tion, minimum, 25th percentile (Q1), median (50th percentile or Q2),
75th percentile (Q3), and maximum of the columns.
1
CS214: Artificial Intelligence Laboratory—Lab 1
4. Calculate and display the following statistical measures for each at-
tribute using the respective functions: mean using ‘mean()‘, median
using ‘median()‘, mode using ‘mode()‘, minimum using ‘min()‘, maxi-
mum using ‘max()‘, and standard deviation using ‘std()‘. Additionally,
compute the quartiles (Q1, Q2, and Q3) using ‘quantile()‘ and the
Interquartile Range (IQR) by subtracting Q1 from Q3.
5. Generate scatter plots to explore the relationships between skin, Age,
BMI, pregs, and plas.
6. Plot histograms for all numerical attributes and the kernel density es-
timation (KDE) curve.
7. Group the data according to the attribute ‘class‘ and bar plots to an-
alyze the distribution of the attributes ‘BMI‘, ‘Age‘, ‘plas‘, and ‘pres‘
for each value of the ‘class‘ variable using the ‘groupby()‘ function.
8. Create boxplots for all attributes to identify outliers and compare dis-
tributions.
Problem Statement 2
A dataset related to red variants of the Portuguese ”Vinho Verde” wine is
given as a CSV file (winequality-red.csv). This dataset contains the values
of different physicochemical tests from each sample of red wine [1]. The orig-
inal goal of the dataset is to model wine quality based on physicochemical
tests. The attributes of the dataset based on physicochemical tests are fixed
acidity, volatile acidity, citric acid, residual sugar, chlorides, free sulfur diox-
ide, total sulfur dioxide, density, pH value, sulphates, alcohol content, and
the last attribute is quality. Each expert graded the wine quality between 0
(very bad) and 10 (very excellent).
Write a Python program to perform the following tasks:
1. Convert the given CSV file to a DataFrame and display the last 10
tuples using the ‘tail()‘ function.
2. Display the structure of the data to provide details about the number of
entries, data types, and memory usage of the dataset using the ‘info()‘
function.
Indian Institute of Technology Dharwad 2
CS214: Artificial Intelligence Laboratory—Lab 1
3. Generate the descriptive statistics for each numerical attribute using
the ‘describe()‘ function to provide the count, mean, standard devia-
tion, minimum, 25th percentile (Q1), median (50th percentile or Q2),
75th percentile (Q3), and maximum of the columns.
4. Calculate and display the following statistical measures for each at-
tribute using the respective functions: mean using ‘mean()‘, median
using ‘median()‘, mode using ‘mode()‘, minimum using ‘min()‘, maxi-
mum using ‘max()‘, and standard deviation using ‘std()‘. Additionally,
compute the quartiles (Q1, Q2, and Q3) using ‘quantile()‘ and the
Interquartile Range (IQR) by subtracting Q1 from Q3.
5. Generate scatter plots to explore the relationships between each at-
tribute citric acid, residual sugar, chlorides, pH, sulphates,
alcohol.
6. Plot histograms for all numerical attributes and the kernel density es-
timation (KDE) curve.
7. Group the dataset by the variable quality and bar plots to analyze
the distribution of the attributes citric acid, residual sugar, free
sulfur dioxide, and total sulfur dioxide for each quality group.
Use the ‘groupby()‘ function to achieve this and visualize the results.
8. Create boxplots for all attributes to identify outliers and compare dis-
tributions.
Note
Kindly upload the completed Jupyter Notebook for each of the
problem statements separately to Moodle for evaluation.
Deliverables
• Python scripts containing the implementation of all the specified tasks.
• Visualizations generated for scatter plots, histograms, bar plots, and
boxplots.
• A summary of findings from the statistical analysis and visualizations.
Indian Institute of Technology Dharwad 3
CS214: Artificial Intelligence Laboratory—Lab 1
References
[1] P. Cortez, A. Cerdeira, F. Almeida, T. Matos, and J. Reis, “Modeling
wine preferences by data mining from physicochemical properties,” De-
cision Support Systems, vol. 47, no. 4, pp. 547–553, 2009.
Indian Institute of Technology Dharwad 4
