2b.
Graphical Representation
Instructions:
Please share your answers filled in-line in the word document. Submit code
separately wherever applicable.
Please ensure you update all the details:
Name: _____________ Batch ID: ___________
Topic: Data Visualization
Guidelines:
1. An assignment submission is considered complete only when the correct and executable code(s) is
submitted along with the documentation explaining the method and results. Failing to submit either
of those will be considered an invalid submission and will not be considered a correct submission.
2. Ensure that you submit your assignments correctly. Resubmission is not allowed.
3. Post the submission you can evaluate your work by referring to the keys provided. (will be available
only post the submission).
Hints: Follow CRISP-ML(Q) methodology steps, where were appropriate.
1. Data Understanding: work on each feature of the dataset to create a data
dictionary as displayed in the image below:
Make a table as shown above and provide information about the features such as its data
type and its relevance to the model building. And if not relevant, provide reasons and a
description of the feature.
Problem Statements:
1. Univariate plots for UNIV data (Plot must have Title, X & Y label)
A) Plot numerical column with 3 different plots ?
B) What are bin parameters? What are the methods to define the number of bins
and bin sizes ?
○ Ans:)
● Number of bins, Bin width, Bin edges, Starting point.
● Number of bins - Square root method , Sturges' formula , Freedman-Diaconis' rule
● Bin width = (max value - min value) / number of bins
C) Why do density plots exceed the range values of the column ?
○ Ans:)
● The density plot can extend beyond the actual values because it is an estimate of the
probability density function of the underlying distribution that the data comes from.
D) Plot categorical columns by taking unique values ?
© 360DigiTMG. All Rights Reserved.
2. Bivariate graphs for UNIV data (Plot must be readable [use rotation], have all labels)
A) Plot 2 numerical columns with scatter plot [use grid] ?
B) 2 Different plots for plotting a numerical column with a categorical column (bar,
line) ?
C) How are bar plots different from histogram?
Ans:)
Bar Plots Histograms
Used to compare the values of Used to show the distribution of a
different categories or groups. single variable.
Can be vertical or horizontal. Can only be vertical.
Bars are separated and do not touch Bars are touching each other and form
each other. a continuous distribution.
Suitable for categorical or discrete Suitable for continuous or numerical
data. data.
X-axis represents categories or groups. X-axis represents the range of the
variable being measured.
Y-axis represents a numerical value. Y-axis represents the frequency or
count of the data in each bin.
3. Plot multivariate graphs (correlation heatmap, pairplot)
A) Plot for only numerical data ?
B) Plot multivariate graphs for both numerical and categorical columns ?
C) What does it mean when a correlation value says 1? When it is negative? When it is
zero?
Ans:)
Correlation value of 1: It means that there is a perfect positive correlation between the two
variables. This means that when one variable increases, the other variable also increases
proportionally, and vice versa.
Correlation value of -1: It means that there is a perfect negative correlation between the two
variables. This means that when one variable increases, the other variable decreases
proportionally, and vice versa.
Correlation value of 0: It means that there is no correlation between the two variables. This
means that there is no relationship between the variables. However, it's important to note that
a correlation of 0 does not necessarily imply independence between the variables, as there may
be other types of relationships that are not captured by correlation.
4. Plot Skewness & Probability distribution for each column of marks data. (Hist, box,
density)
© 360DigiTMG. All Rights Reserved.
A) What is normally distributed and What will be the relationship between mean,
median & mode ?
Ans:)
● Math column is normally distributed. Generally, Mean = Median = Mode when data is
normally distributed.
B) Which data variables are positively skewed and What will be the relationship
between mean, median & mode ?
Ans:
● Science column is positively distributed. Generally, Mean > Median > Mode when data is
positively distributed.
C) What are negatively skewed/distributed and What will be the relationship
between mean, median & mode
Ans:)
● Social studies column is positively distributed Generally Mean < Median < Mode when
data is negatively distributed.
D) What are the distinctive differences between skewness and distribution?
Ans:)
Distinctive differences between skewness and distribution:
● Distribution refers to the shape or spread of the data points in a dataset, while
skewness refers to the degree and direction of the asymmetry of the
distribution.
● Skewness measures how far the data is spread out on one side of the mean
compared to the other side, while distribution refers to the values of the data
points themselves.
● A distribution can be symmetrical, positively skewed, or negatively skewed, while
skewness can be positive, negative, or zero.
● Skewness can be affected by extreme values, while distribution is not.
© 360DigiTMG. All Rights Reserved.
