Data Storytelling with Altair and AI

brief contents
contents
preface
acknowledgments
about this book
	Who should read this book
	How this book is organized: A road map
	About the code
	liveBook discussion forum
about the author
about the cover illustration
Part 1—Introducing Altair and generative AI to data storytelling
	1 Introducing data storytelling
		1.1 The art of data storytelling
			1.1.1 Why should you use data storytelling?
			1.1.2 What problems can data storytelling solve?
			1.1.3 What are the challenges of data storytelling?
		1.2 Why should you use Python Altair and generative AI for data storytelling?
			1.2.1 The benefits of using Python in all the steps of the data science project life cycle
			1.2.2 The benefits of using generative AI for data storytelling
		1.3 When Altair and generative AI tools are not useful for data storytelling
		1.4 Using the data, information, knowledge, wisdom pyramid for data storytelling
			1.4.1 From data to information
			1.4.2 From information to knowledge
			1.4.3 From knowledge to wisdom
		Summary
		References
	2 Running your first data story in Altair and GitHub Copilot
		2.1 Introducing Altair
			2.1.1 Chart
			2.1.2 Mark
			2.1.3 Encodings
		2.2 Use case: Describing the scenario
			2.2.1 The dataset
			2.2.2 Data exploration
		2.3 First approach: Altair
			2.3.1 From data to information
			2.3.2 From information to knowledge
			2.3.3 From knowledge to wisdom
			2.3.4 Comparing Altair and Matplotlib
		2.4 A second approach: Copilot
			2.4.1 Loading and cleaning the dataset
			2.4.2 Calculating the percentage increase
			2.4.3 Plotting the basic chart in Altair
			2.4.4 Enriching the chart
		Summary
		References
			EDA
			Tools and libraries
			Other
	3 Reviewing the basic concepts of Altair
		3.1 Vega and Vega-Lite
			3.1.1 Vega
			3.1.2 Vega-Lite
			3.1.3 How to render a Vega or Vega-Lite visualization
		3.2 The basic components of an Altair chart
			3.2.1 Encodings
			3.2.2 Marks
			3.2.3 Conditions
			3.2.4 Compound charts
			3.2.5 Interactivity
			3.2.6 Configurations
		3.3 Case study
			3.3.1 From data to information
			3.3.2 From information to knowledge
			3.3.3 From knowledge to wisdom
		Summary
		References
			Data visualization
			Vega, Vega-Lite, and D3.js
			Other
	4 Generative AI tools for data storytelling
		4.1 Generative AI tools: On the giants’ shoulders
			4.1.1 What is artificial intelligence?
			4.1.2 What is machine learning?
			4.1.3 What is deep learning?
			4.1.4 What is generative AI?
			4.1.5 Generative AI tools landscape
		4.2 The basic structure of a ChatGPT prompt
			4.2.1 Defining the task
			4.2.2 Acting as a role
			4.2.3 Tailoring to an audience
		4.3 The basic structure of a DALL-E prompt
			4.3.1 Subject
			4.3.2 Style
			4.3.3 The Edit Image tool
		4.4 Using Copilot to build the components of an Altair chart
			4.4.1 Prerequisites
			4.4.2 Marks
			4.4.3 Encodings
			4.4.4 Conditions
			4.4.5 Compound charts
			4.4.6 Interactivity
		4.5 Case study: Your training team
			4.5.1 Turning data into information
			4.5.2 Turning information into knowledge
			4.5.3 Turning knowledge into wisdom
		Summary
		References
			Prompt engineering
			Other
Part 2—Using the DIKW pyramid for data storytelling
	5 Crafting a data story using the DIKW pyramid
		5.1 Breaking the ice: The homelessness tale
			5.1.1 What was wrong with the chart?
			5.1.2 What was wrong with the presentation?
		5.2 Uncovering the narrative: What a data story is
			5.2.1 Using the DIKW pyramid to streamline a data story
			5.2.2 DIKW in action: Completing the homelessness tale
		5.3 Incorporating generative AI into the DIKW pyramid
		5.4 Behind the scenes: The homelessness tale
			5.4.1 Creating a compelling subtitle
			5.4.2 Generating images
		5.5 Another example: Fake news
			5.5.1 From data to information
			5.5.2 From information to knowledge
			5.5.3 From knowledge to wisdom
		Summary
		References
	6 From data to information: Extracting insights
		6.1 An intuitive approach to extract insights
			6.1.1 Connection strategy
			6.1.2 Coincidence strategy
			6.1.3 Curiosity
			6.1.4 Contradictions
		6.2 Choosing the characters of your story
		6.3 Choosing the right chart
			6.3.1 The cooking charts family
			6.3.2 The bar charts family
			6.3.3 The line charts family
			6.3.4 The geographical map family
			6.3.5 Dot charts family
		6.4 Case study: Salmon aquaculture
		Summary
		References
	7 From information to knowledge: Building textual context
		7.1 Introducing context
		7.2 Calibrating the story to the audience
			7.2.1 General public
			7.2.2 Executives
			7.2.3 Professionals
		7.3 Using ChatGPT for commentaries and annotations
			7.3.1 Describing the topic
			7.3.2 Describing the type
			7.3.3 Setting custom instructions
		7.4 Using large language models for context
			7.4.1 Fine-tuning
			7.4.2 Retrieval augmented generation
		7.5 Case study: From information to knowledge (part 1)
			7.5.1 Tailoring the chart to the audience
			7.5.2 Using RAG to add a commentary
			7.5.3 Highlighting the period of decrease in sales
			7.5.4 Exercise
		Summary
		References
			Embeddings
			Fine-tuning
			LangChain
			LLM
			RAG
			Thinking for the audience
			Transformers
	8 From information to knowledge: Building the visual context
		8.1 Emotions: The foundations of visual context
		8.2 Color
			8.2.1 Setting colors in Altair
			8.2.2 Exercise: Setting colors
		8.3 Size
			8.3.1 Font size
			8.3.2 Chart size
			8.3.3 Exercise: Setting size
		8.4 Interaction
			8.4.1 Tooltip
			8.4.2 Slider
			8.4.3 Drop-down menu
			8.4.4 Exercise: Setting interactivity
		8.5 Using DALL-E for images
			8.5.1 Adding emotions
			8.5.2 Generating consistent images
			8.5.3 Exercise: Generating images
		8.6 Strategic placement of context
			8.6.1 Top placement
			8.6.2 Left placement
			8.6.3 Within placement
		8.7 Case study: From information to knowledge (part 2)
			8.7.1 Setting a negative mood
			8.7.2 Setting a positive mood
			8.7.3 Exercise: Making the chart interactive
		Summary
		References
			Using emotions for communication
			Images
	9 From knowledge to wisdom: Adding next steps
		9.1 Introducing wisdom
			9.1.1 Transforming knowledge into wisdom: Next steps
			9.1.2 Using ChatGPT as a source of experience
			9.1.3 Good judgment: Anchoring the action to an ethical framework
		9.2 Case studies
			9.2.1 Chapter 1: The dogs and cats campaign
			9.2.2 Chapter 2: The tourist arrivals
			9.2.3 Chapter 3: Population in North America
			9.2.4 Chapter 4: Sport disciplines
			9.2.5 Chapter 5: Homelessness
			9.2.6 Chapter 5: Fake news
			9.2.7 Chapters 6–8: The salmon aquaculture case study
		9.3 Strategic placement of next steps
			9.3.1 Title placement
			9.3.2 Right placement
			9.3.3 Below placement
		Summary
		References
Part 3—Delivering the data story
	10 Common issues while using generative AI
		10.1 Hallucination, bias, and copyright
			10.1.1 AI hallucinations
			10.1.2 Bias
			10.1.3 Copyright
		10.2 Guidelines for using generative AI
		10.3 Crediting the sources
			10.3.1 Under the title or subtitle
			10.3.2 Under the main chart
			10.3.3 Under the next steps
			10.3.4 Sideways
			10.3.5 Implementing credits in Altair
		Summary
		References
			Generative AI issues
			Ethics and AI
	11 Publishing the data story
		11.1 Exporting the story
		11.2 Publishing the story over the web: Streamlit
		11.3 Tableau
		11.4 Power BI
		11.5 Comet
		11.6 Presenting through slides
		11.7 Final thoughts
		Summary
		References
appendix A—Technical requirements
	A.1 Cloning the GitHub repository
		A.1.1 Using the terminal
		A.1.2 Using GitHub Desktop
	A.2 Installing the Python packages
	A.3 Installing GitHub Copilot
	A.4 Configuring ChatGPT
	A.5 Installing Open AI API
	A.6 Installing LangChain
	A.7 Installing Chroma
	A.8 Configuring DALL-E
appendix B—Python pandas DataFrame
	B.1 An overview of pandas DataFrame
		B.1.1 Building from a dictionary
		B.1.2 Building from a CSV file
	B.2 dt
	B.3 groupby()
	B.4 isnull()
	B.5 melt()
	B.6 unique()
appendix C—Other chart types
	C.1 Donut chart
	C.2 Bar charts family
		C.2.1 Column chart
		C.2.2 Column chart with multiple series
		C.2.3 Pyramid chart
		C.2.4 Stacked column chart
		C.2.5 100% stacked column chart
		C.2.6 Histograms
	C.3 Line charts family
		C.3.1 Area chart
		C.3.2 Slope chart
		C.3.3 Dumbbell chart
index
	Symbols
	Numerics
	A
	B
	C
	D
	E
	F
	G
	H
	I
	J
	K
	L
	M
	N
	O
	P
	Q
	R
	S
	T
	U
	V
	W
	X
	Y