Data Science for Web3

✍ Scribed by Gabriela Castillo Areco

• Publisher: Packt Publishing Pvt. Ltd.
• Year: 2023
• Tongue: English
• Leaves: 848
• Category: Library

✦ Synopsis

​From zero to hero with blockchain data. Learn how to travel a business data science pipeline and build real world solutions for Web 3

Key Features
​Get to grips with the blockchain´s analytics fundamentals
​Discover real world solutions by modeling blockchain data and learn to present/ deploy your projects
​Showcase your work and learn to build experience to launch your next opportunity in Web 3

Book Description
​Data is the new Oil and Web 3 is starting to generate lots of it. This book aims to be a step-by-step guide for a data science project with practical examples, explanations, and ideas for portfolio development. ​In four main blocks, independent from each other, you’ll discover the industry best practices, tools, and sources needed to navigate your way as a professional of data in Web 3. Start by understanding blockchain key concepts and the essential data science tools needed to work in Web 3 projects. ​Read into the next chapters to explore the main sources of data that can help cope with those industry challenge​s​, decoding smart contracts and building datasets. ​Disentangle feature engineering questions that may arise when dealing with blockchain data before diving in deeper into machine learning use cases where crypto data is being used. The book includes real world examples based on interviewed executives from renowned companies that are facing technical challenges when using data to foster innovation in the Web 3 environment.

​Lastly, we will learn how to build experience ​in​ crypto data that you will able to demonstrate to recruiters in job interviews, academic purposes or potential clients. By the end of this book, you will be able to build complete data science projects using blockchain data.

What you will learn
Discover onchain and offchain data and build datasets for ERC 721 tokens
Leverage data science to answer industry-specific business questions in Web 3
Learn where and how to showcase your work to land a Web 3 job
Develop a portfolio showcasing sentiment analysis and crypto art generation projects

Who This Book Is For
​If you are a Data Scientist, Blockchain developer, Web 3 Engineer you will learn how to extract data from the Web 3 ecosystem. If you are a business related professional additionally you will learn to use data science tools in the context of blockchain data. Some information of Data science would be needed to get the best of this book.

✦ Table of Contents

Data Science for Web3
Foreword
Contributors
About the author
About the reviewer
Preface
Who this book is for
What this book covers
To get the most out of this book
Download the example code files
Conventions used
Get in touch
Share Your Thoughts
Download a free PDF copy of this book
Part 1 Web3 Data Analysis Basics
1
Where Data and Web3 Meet
Technical requirements
Exploring the data ingredients
Understanding the blockchain ingredients
Three generations of blockchain
Introducing the blockchain ingredients
Making the first transaction
Approaching Web3 industry metrics
Block height
Time
Tokenomics
Total Value Locked (TVL)
Total market cap
Data quality challenges
Data standards challenges
Retail
Confirmations
NFT Floor Price
The concept of “lost”
A brief overview of APIs
Summary
Further reading
2
Working with On-Chain Data
Technical requirements
Dissecting a transaction
Nonce
Gas price
Gas limit
Recipient
Sender
Value
Input data
V,R,S
Transaction receipt
Status
Gas used and Cumulative gas used
Logs
Dissecting a block
Exploring state data
Reviewing data sources
Block explorers
Infura
Moralis
GetBlock
Dune
Covalent
Flipside
The Graph
Google BigQuery
Summary
Further reading
3
Working with Off-Chain Data
Technical requirements
Introductory example – listing data sources
Adding prices to our dataset
CoinGecko
CoinMarketCap
Binance
Oracles – Chainlink
OHLC – Kraken
Final thoughts on prices
Adding news to our dataset
Adding social networks to our dataset
X (formerly Twitter)
Summary
Further reading
4
Exploring the Digital Uniqueness of NFTs – Games, Art, and Identity
Enabling unique asset tracking on the blockchain using NFT
The business requests
The technical solution
Blockchain gaming – the GameFi proposal
Introduction to the business landscape
Analytics
Identity in the blockchain
Introduction to the business landscape
Analytics
Redefining the art business with blockchain
Introduction to the business landscape
Data extraction
Floor price and wash trading
A word on anti-money laundering (AML) practices
Summary
Further reading
5
Exploring Analytics on DeFi
Technical requirements
Stablecoins and other tokens
Understanding tokens, native assets, and the ERC-20 data structure
Hands-on example
Understanding DEX
Hands-on example – pools and AMM
DEX aggregators
Lending and borrowing services on Web3
Flash loans
A note on protocol bad debt
Multichain protocols and cross-chain bridges
Hands-on example – Hop bridge
Summary
Further reading
Part 2 Web3 Machine Learning Cases
6
Preparing and Exploring Our Data
Technical requirements
Data preparation
Hex values
Checksum
Decimal treatment
From Unix timestamps to datetime formats
Evolution of smart contracts
Exploratory Data Analysis
Summarizing data
Outlier detection
Summary
Further reading
7
A Primer on Machine Learning and Deep Learning
Technical requirements
Introducing machine learning
Building a machine learning pipeline
Model
Training
Underfitting and overfitting
Prediction and evaluation
Introducing deep learning
Model preparation
Model building
Training and evaluating a model
Summary
Further reading
8
Sentiment Analysis – NLP and Crypto News
Technical requirements
Example datasets
Building our pipeline
Preparation
Model building
Training and evaluation
ChatGPT integration
Summary
Further reading
9
Generative Art for NFTs
Technical requirements
Creating with colors – colorizing
Hands-on Style2Paints
Theory
A note on training datasets
Creating with style – style transfer
Preparation
Model building
Training and inference
Creating with prompts – text to image
DALL.E 2
Stable Diffusion
Midjourney
Leonardo.Ai
Minting an NFT collection
Summary
Further reading
10
A Primer on Security and Fraud Detection
Technical requirements
A primer on illicit activity on Ethereum
Preprocessing
Training the model
Evaluating the results
Presenting results
Summary
Further reading
11
Price Prediction with Time Series
Technical requirements
A primer on time series
Exploring the dataset
Discussing traditional pipelines
Preprocessing
Modeling – ARIMA/SARIMAX and Auto ARIMA
Auto ARIMA
Adding exogenous variables
Using a neural network – LSTM
Preprocessing
Model building
Training and evaluation
Summary
Further reading
12
Marketing Discovery with Graphs
Technical requirements
A primer on graphs
Types of graphs
Graph properties
The dataset
Node classification
Preparation
Modeling
Training and evaluation
Summary
Further reading
Part 3 Appendix
13
Building Experience with Crypto Data – BUIDL
Showcasing your work – portfolio
Looking for a job
Preparing for a job interview
Importance of soft skills
Where to study
Summary
Further reading
14
Interviews with Web3 Data Leaders
Hildebert Moulié (aka hildobby)
Jackie Zhang
Marina Ghosh
Professor One Digit
Appendix 1
Google Colaboratory
Anaconda
Ganache
Infura
Appendix 2
Appendix 3
Index
Why subscribe?
Other Books You May Enjoy
Packt is searching for authors like you
Share Your Thoughts
Download a free PDF copy of this book