Book Image

Principles of Data Science - Third Edition

By : Sinan Ozdemir
Book Image

Principles of Data Science - Third Edition

By: Sinan Ozdemir

Overview of this book

Principles of Data Science bridges mathematics, programming, and business analysis, empowering you to confidently pose and address complex data questions and construct effective machine learning pipelines. This book will equip you with the tools to transform abstract concepts and raw statistics into actionable insights. Starting with cleaning and preparation, you’ll explore effective data mining strategies and techniques before moving on to building a holistic picture of how every piece of the data science puzzle fits together. Throughout the book, you’ll discover statistical models with which you can control and navigate even the densest or the sparsest of datasets and learn how to create powerful visualizations that communicate the stories hidden in your data. With a focus on application, this edition covers advanced transfer learning and pre-trained models for NLP and vision tasks. You’ll get to grips with advanced techniques for mitigating algorithmic bias in data as well as models and addressing model and data drift. Finally, you’ll explore medium-level data governance, including data provenance, privacy, and deletion request handling. By the end of this data science book, you'll have learned the fundamentals of computational mathematics and statistics, all while navigating the intricacies of modern ML and large pre-trained models like GPT and BERT.

Related Content you might be interested in

Current Title:

Small book image
Principles of Data Science - Third Edition
Sinan Ozdemir
Jan, 2024 | 326 pages
Small book image
Feature Engineering Made Easy
Divya Susarla | Sinan Ozdemir
Jan, 2018 | 316 pages
Small book image
Cracking the Data Science Interview
Leondra R Gonzalez | Aaren Stubberfield
Feb, 2024 | 404 pages
Small book image
Responsible AI in the Enterprise
Adnan Masood | Heather Dawe
Jul, 2023 | 318 pages
Table of Contents (18 chapters)
Preface
Preface
Who is this book 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
Free Chapter
1
Chapter 1: Data Science Terminology
Chapter 1: Data Science Terminology
What is data science?
The data science Venn diagram
Some more terminology
Data science case studies
Summary
2
Chapter 2: Types of Data
Chapter 2: Types of Data
Structured versus unstructured data
The four levels of data
Summary
Questions and answers
3
Chapter 3: The Five Steps of Data Science
Chapter 3: The Five Steps of Data Science
Introduction to data science
Exploring the data
Summary
4
Chapter 4: Basic Mathematics
Chapter 4: Basic Mathematics
Basic symbols and terminology
Linear algebra
Summary
5
Chapter 5: Impossible or Improbable – A Gentle Introduction to Probability
Chapter 5: Impossible or Improbable – A Gentle Introduction to Probability
Basic definitions
Bayesian versus frequentist
How to utilize the rules of probability
Introduction to binary classifiers
Summary
6
Chapter 6: Advanced Probability
Chapter 6: Advanced Probability
Bayesian ideas revisited
Random variables
Summary
7
Chapter 7: What Are the Chances? An Introduction to Statistics
Chapter 7: What Are the Chances? An Introduction to Statistics
What are statistics?
How do we obtain and sample data?
How do we measure statistics?
The empirical rule
Summary
8
Chapter 8: Advanced Statistics
Chapter 8: Advanced Statistics
Understanding point estimates
Sampling distributions
Confidence intervals
Hypothesis tests
Summary
9
Chapter 9: Communicating Data
Chapter 9: Communicating Data
Why does communication matter?
Identifying effective visualizations
When graphs and statistics lie
Verbal communication
Summary
10
Chapter 10: How to Tell if Your Toaster is Learning – Machine Learning Essentials
Chapter 10: How to Tell if Your Toaster is Learning – Machine Learning Essentials
Introducing ML
Types of ML
Predicting continuous variables with linear regression
Summary
11
Chapter 11: Predictions Don’t Grow on Trees, or Do They?
Chapter 11: Predictions Don’t Grow on Trees, or Do They?
Performing naïve Bayes classification
Understanding decision trees
Diving deep into UL
Feature extraction and PCA
Summary
12
Chapter 12: Introduction to Transfer Learning and Pre-Trained Models
Chapter 12: Introduction to Transfer Learning and Pre-Trained Models
Understanding pre-trained models
Different types of TL
TL with BERT and GPT
Summary
13
Chapter 13: Mitigating Algorithmic Bias and Tackling Model and Data Drift
Chapter 13: Mitigating Algorithmic Bias and Tackling Model and Data Drift
Understanding algorithmic bias
Sources of algorithmic bias
Measuring bias
Consequences of unaddressed bias and the importance of fairness
Mitigating algorithmic bias
Bias in LLMs
Emerging techniques in bias and fairness in ML
Understanding model drift and decay
Mitigating drift
Summary
14
Chapter 14: AI Governance
Chapter 14: AI Governance
Mastering data governance
Navigating the intricacy and the anatomy of ML governance
A guide to architectural governance
Summary
15
Chapter 15: Navigating Real-World Data Science Case Studies in Action
Chapter 15: Navigating Real-World Data Science Case Studies in Action
Introduction to the COMPAS dataset case study
Text embeddings using pretrainedmodels and OpenAI
Summary
16
Index
Index
Why subscribe?
17
Other Books You May Enjoy
Other Books You May Enjoy
Packt is searching for authors like you
Share Your Thoughts
Download a free PDF copy of this book

Summary

In this chapter, we explored basic data science terminology and saw how even the term “data science” can be fraught with ambiguity and misconception. We also learned that coding, math, and domain expertise are the fundamental building blocks of data science. As we seek new and innovative ways to discover data trends, a beast lurks in the shadows. I’m not talking about the learning curve of mathematics or programming, nor am I referring to the surplus of data. The Industrial Age left us with an ongoing battle against pollution. The subsequent Information Age left behind a trail of big data. So, what dangers might the Data Age bring us?

The Data Age can lead to something much more sinister – the dehumanization of the individual through mass data and the rise of automated bias in machine learning systems.

More and more people are jumping headfirst into the field of data science, most with no prior experience in math or computer science, which, on the surface, is great. The average data scientist has access to millions of dating profiles’ data, tweets, online reviews, and much more to jump-start their education. However, if you jump into data science without the proper exposure to theory or coding practices, and without respect for the domain you are working in, you face the risk of oversimplifying the very phenomenon you are trying to model.

Now, it’s time to begin. In the next chapter, we will explore the different types of data that exist in the world, ranging from free-form text to highly structured row/column files. We will also look at the mathematical operations that are allowed for different types of data, as well as deduce insights based on the form the data comes in.

Previous Section
End of Chapter 1
Next Chapter