Introduction

Design the output (Video 1)

Video summary

This video explains how to design the output of data analysis, focusing on data visualization
and storytelling. It covers general-purpose and specialized tools for creating effective
visualizations.

Highlights:

• 00:133 Introduction to designing the output

o Importance of showing and telling data
o Focus on data visualization and storytelling
o Overview of the module’s content
• 00:424 Data visualization and dashboards
o Role of data visualization in design
o Common use of dashboards
o Tools for creating dashboards
• 01:405 General-purpose tools
o Excel for unusual visualizations
o BI tools like Google Data Studio, Power BI, and Tableau
o Familiarity with these tools is beneficial
• 02:116 Specialized tools
o Excel for non-standard charts
o Flourish Studio for animated charts
o Kumo for network visualizations
• 03:217 Geographic visualizations
o Importance of processing and color-coding data
o Use of tools like UGis
o Introduction to a few specialized tools
• 04:018 Choosing the right tool
o Importance of familiarity with tools
o Tools as a means to an end
o Designing outputs for better audience understanding

Visualizing forecasts with Excel (Video 2)

Video summary

This video demonstrates how to create visualizations in Excel to explore time series data,
focusing on securities like currencies, indices, and commodities. It covers creating pivot
tables, sparklines, and correlation matrices to analyze trends, forecast values, and identify
correlations.

Highlights:

• 00:003 Introduction to data visualization in Excel

o Exploring time series data
o Securities include currencies, indices, and commodities
o Data range from August to September
• 01:184 Analyzing trends and fluctuations
o Identifying trends in securities
o Comparing indices and currencies
o Forecasting future values
• 03:385 Creating visualizations
o Using pivot tables
o Adding sparklines for trends
o Calculating averages and forecasts
• 10:536 Correlation analysis
o Measuring correlation between securities
o Creating scatter plots and trendlines
o Using correlation coefficients
• 18:017 Building a correlation matrix
o Using the data analysis tool pack
o Applying conditional formatting
o Identifying patterns and outliers

• Excel File

Visualizing animated data with PowerPoint (Video 3)

Video summary

This video tutorial demonstrates how to create a bar chart race in PowerPoint, using historical
GDP data of various countries. The process involves data preparation in Excel, creating and
formatting shapes in PowerPoint, and using the Morph transition for animation.

Highlights:

• 00:003 Introduction to GDP data

o Overview of GDP rankings from 2000 to 2020
o Explanation of the bar chart race concept
o Introduction to the tutorial’s objectives
• 01:044 Data preparation in Excel
 o Copying GDP data from Wikipedia
o Cleaning and organizing the data
o Creating a pivot table for visualization
• 05:045 Creating the bar chart race in PowerPoint
o Setting up the PowerPoint slide
o Drawing shapes for each country
o Inputting GDP values and country names
• 12:006 Animating the bar chart race
o Using the Morph transition for smooth animation
o Adjusting the speed and alignment of transitions
o Ensuring accurate representation of data changes
• 17:007 Final adjustments and improvements
o Enhancing visual impact with faster transitions
o Adding and removing countries dynamically
o Final review of the animated bar chart race

I hope this helps! If you have any questions or need further assistance, feel free to ask.

• How to make a bar chart race in PowerPoint

Visualizing animated data with Flourish (Video 4)

Here are the key highlights from the video “Flourish ‘Tea Breaks’: Animation”:

• Introduction: Richard, the host, welcomes attendees and introduces the topic of
animation in Flourish.
• Animation Principles: Emphasis on using animation to both engage and inform, with
a focus on object constancy to reduce cognitive load.
• Templates Overview: Demonstrations of animation settings in various Flourish
templates like line chart race, bar chart race, and linebar pi.
• Practical Tips: Guidance on adjusting animation settings for different data sets and
ensuring smooth transitions in data storytelling.

Visualizing network data with Kumu (Video 5)

Video summary

This video tutorial introduces Kumu, a tool for visualizing complex relationships in data,
such as social network analysis. The tutorial focuses on preparing data for Kumu using IMDb
data of actors in Indian movies, demonstrating how to create and visualize a network of
actors who have worked together.

Highlights:

• 00:003 Introduction to Kumu

 o Visualizes complex relationships
o Useful for social network analysis
o Helps understand and ask questions to the network
• 01:194 Data preparation
o Using IMDb data of Indian movie actors
o Filtering data to focus on movies
o Creating a matrix of actor relationships
• 04:015 Matrix creation
o Matrix multiplication for actor connections
o Sparse matrix representation
o Efficient storage using compressed sparse row format
• 09:016 Matrix conversion and upload
o Converting matrix to sparse format
o Uploading data to Kumu
o Visualizing actor networks in Kumu
• 12:047 Network visualization
o Exploring actor connections
o Identifying clusters and links
o Example of South Indian superstar’s network

• Kumu
• IMDB data
• Jupyter Notebook

Optional : Network analysis – filtering by year (Video 6)

Video summary

This video demonstrates how to filter data by year and language in network analysis using
Python. It covers importing packages, handling errors, and filtering data frames.

Highlights:

• 00:033 Introduction and setup

o Execute code block by block
o Edit code to filter by year and language
o Import necessary packages
• 00:244 Handling errors and package versions
o Address network adjacency function error
o Install specific version of SK Network
o Use version 0.24 for compatibility
• 02:315 Filtering data by year
o Read and filter title data frame
o Convert start year to integer
o Filter movies released after a specific year
• 08:116 Boolean filtering and data manipulation
o Use Boolean vectors for filtering
o Handle string and integer conversions
o Drop rows with invalid data
• 31:007 Filtering data by language
 o Filter data frame by language
o Use language option in function calls
o Check unique values in region data frame

Visualizing charts with Excel (Video 7)

PART 1Video summary

This video by Anand S discusses the importance of visualizing machine learning models and
provides practical examples of how to interpret and explain complex models using
visualization techniques.

Highlights:

• 00:013 Introduction to black box models

o Explanation of black box models
o Example of a telecom company predicting customer churn
o Challenges with black box models in practical applications
• 03:014 Improving model accuracy
o Use of decision trees, support vector machines, and neural networks
o Achieving significant cost reduction in customer retention
o Issues with model explainability and acceptance
• 05:325 Understanding neural networks
o Forecasting agricultural prices using neural networks
o Achieving high accuracy with neural network models
o Difficulty in explaining neural network models to clients
• 07:426 K-means clustering example
o Clustering census data to identify similar districts
o Visualizing clusters on a map
o Practical applications of clustering results
• 13:317 Interpreting clustering results
o Identifying key characteristics of clusters
o Using visualization tools to understand clusters
o Practical implications for marketing and content strategy
• 21:018 Moving up and down the ladder of abstraction
o Explanation of the ladder of abstraction concept
o Using kinesthetic understanding to interpret models
o Practical example with a self-driving car model

Display animated data using Flourish (Video 8)

PART 1Video summary

This video provides an overview of the different animation options available in Flourish,
focusing on how to use them effectively to engage and inform your audience. The host,
Richard, explains key principles and settings for various templates, including line chart races,
bar chart races, and the linebar pie template. He emphasizes the importance of using
animation to enhance data storytelling and reduce cognitive load.

Highlights:

• 00:013 Introduction and Overview

o Host Richard introduces the webinar
o Announces bi-weekly schedule for 2021
o Emphasizes engaging and informative animations
• 03:064 Line Chart Race Settings
o Explanation of animation duration and mode duration
o Demonstration with a simple line chart race example
o Adjusting settings for different data points
• 09:205 Bar Chart Race Settings
o Duration of the timeline and animation speed
o Example with Brazil’s data
o Adjusting settings for smooth transitions
• 11:126 Linebar Pie Template
o Drawing and morphing animations
o Example with UK energy trends data
o Controlling animation speed and turning off animations
• 16:017 Animating Series with Different Names
o Importance of consistent series names
o Examples of misleading and effective animations
o Adjusting settings for better data storytelling

Google Charts & Google Data Studio

• Short intro
• Google Data Studio: Google Data Studio Tutorial for Beginners🔥
• Google Chart 1: How To Create Chart Or Graph On HTML CSS Website | Google Charts
Tutorial
• Google Chart 2: Using Google Charts with googleVis

Actor network visualization (Video 9)

Summary:

Imagine if Govinda announces that his next movie would be with Angelina Jolie, and the movie is

titled "Jolie No. 1" Imagine further, that he announced that a co-star connected them, but declined

to say who. Who might this co-star be? The Gramener team explores the social network of actors

and finds the most likely chain of co-actors that can connect Govinda to Angelina -- and who the

critical link in that chain is.

Find the shortest path between Govinda & Angelina Jolie using IMDb data using
Python: networkx or scikit-network.

• Notebook: How this video was created

• The data used to visualize the network
• The shortest path between actors
• IMDB data
• Codebase