Network Graph

Network Graph : Motivation

Given a table showing the friendships between people in a group, you are
assigned to split that group into groups of two.
However, it is difficult to do so by looking at the table.
Network Graph: Motivation
Wouldn’t it be nice if you could visualize their connections using an
interactive network graph like below?
 Network Graph: Pyvis
 What is Pyvis?
Pyvis is a Python library that allows you to create interactive network graphs
in a few lines of code.
To install pyvis, type:
pip install pyvis
 What is a Tree map?
 What is a Tree map?
A tree map is a special type of chart for visualization using a set of nested
rectangles of categorical data that is preferably hierarchical.

 In Hierarchical data, the categories or items share parent-child type

relationships in an overall tree structure.

 The simplest example of this type of data structure can be seen in a

company where all individuals and their designations within teams could be
grouped under one entity i.e., the company itself
 What is a Tree map?
 When to use a Tree map?
These are some key points to consider before using tree maps for
visualization.
Tree maps work well when there is a clear ‘Part-to-whole’ relationship
amongst multiple categories present in the data.
Hierarchical Data is needed. This indicates that the data could be arranged
in branches and sub-branches.
The focus is not on precise comparisons between categories but rather on
spotting the key factors/trends or patterns.
 Benefits of using a Tree map?
 Benefits of using a Tree map:
 Space constraint: There is a large amount of hierarchical data that needs to
be visualized in a smaller space.
 Easier to read: When compared to a circular multi-level pie chart, the tree
map is easier to read due to its linear visual appearance.
 Quickly spot patterns: Since each group is represented by a rectangle and
the area of this rectangle is always proportional to its value, trends and
patterns (similarities and anomalies) are quickly visible in tree maps.
Real-world use cases for Tree map Charts?
 1. Displaying region-wise customer complaints about a product
 Suppose there are 10 different types of complaints (assume these are
denoted as C1 to C10) about a product and the company wants to visualize
which complaints are relevant to a region then in such a case a tree map
could be used. Here, it can be clearly seen how different regions have
specific types of user complaints.
Real-world use cases for Tree map Charts?
2. Showcasing category-wise product availability like mobile phones
Let us assume that there are four categories of mobile phones with their
market share percentages i.e., Low-end (up to 10,000 INR – 15%), Mid-Range
(10,000-25000 INR- 55%), Premium (above 25,000 to 50,000 INR-25%), and
Top-end (above 50,000 INR-10%).
From this tree map, we can gauge that there is a bigger demand and market
for Mid-Range phones while there are limited phones available in the Top-End
category.
Real-world use cases for Tree map Charts?
3. Explore customer segmentation for a product
Usually, companies for apparel or personal products divide their customers
based on their age. This way they can categorize their products and the
product variants separately for each age group.
In the case of this tree map, the company could decide whether to launch
more products for particular customer segments based on the distribution.
 Challenges associated with a Tree map
Tree maps also come with a set of limitations as outlined below-
– Tree maps built with large data points on a single level could be hard to read
as well as print for reporting purposes.
– Sometimes, additional sorting might be required to understand the data
better. However, all the rectangles are automatically ordered within the parent
node by area.
– With too many categories and colors to represent these, the tree map
becomes overwhelming for the reader.
– Tree maps become ineffective for datasets with balanced trees i.e., when
items are of a similar value. In these cases, the main purpose of a tree map of
highlighting the largest item in a given category becomes impossible.
 Dendrograms in Python
A dendrogram is a diagram that depicts a tree.
The create_dendrogram figure factory conducts hierarchical clustering on data
and depicts the resultant tree.
Distances between clusters are represented by the values on the tree depth axis.
Dendrogram plots are often used in computational biology to depict gene or
sample grouping, occasionally in the margins of heatmaps.
Hierarchical clustering produces dendrograms as an output. Many people claim
that dendrograms of this type may be used to determine the number of clusters. 
 Dendrograms in Python
Wholesale Customer Segmentation Problem using Hierarchical Clustering
We will be working on a wholesale customer segmentation problem. You
can download the dataset using this link.
The data is hosted on the UCI Machine Learning repository. The aim of this
problem is to segment the clients of a wholesale distributor based on their
annual spending on diverse product categories, like milk, grocery, region, etc.
 Scree Plot in Python
How to Create a Scree Plot in Python
Principal components analysis (PCA) is an unsupervised machine learning
technique that finds principal components (linear combinations of the
predictor variables) that explain a large portion of the variation in a dataset.

When we perform PCA, we’re interested in understanding what percentage

of the total variation in the dataset can be explained by each principal
component.
One of the easiest ways to visualize the percentage of variation explained
by each principal component is to create a scree plot.