7 Books Every Data Science Student Should Read

It is always helpful for students to read books, articles, blog posts, and anything else outside of formal textbook assignments to round out an education. Taking the initiative to read tangential works in data science is particularly helpful for students taking on this exciting and fast-growing interdisciplinary field.

Near the end of 2020, Niti Sharma from The Next Tech wrote:

“Data science is one of the hottest industries these days, given the massive amounts of data flowing into companies of all sizes and across all sectors.”

If you want to enter this competitive field and plan to pursue your master’s degree in data science at a leading university, you might want to enhance your learning by reading everything relevant you can before, during, and after you earn your degree.

If you are scratching your head, wondering “What book should I start with for data science?” we want to share the following seven books we believe are invaluable in the field and will boost your core knowledge and confidence in your studies as you look to your future career.

1. Introduction to Machine Learning with Python: A Guide for Data Scientists

Written by Andreas Müller, who received his PhD in machine learning from the University of Bonn, Introduction to Machine Learning with Python: A Guide for Data Scientists is essential for students who want to ultimately pursue machine learning.

The book shares various machine learning algorithms and offers an easy-to-follow discussion of how they work without going into intricate mathematical details. It is an ideal introduction to machine learning, allowing people new to the field to get their feet wet without getting bogged down with the theory or mathematics underlying the algorithms.

2. R for Data Science

Per the R Project, R is the “language and environment for statistical computing and graphics.” But there is more to it than that, and this book will offers deeper details. R for Data Science, written by Hadley Wickham and Garrett Grolemund, serves as a primer in helping students learn to do data science with R. With R for Data Science, you will learn:

• How to do data science with R
• How to get your data into R
• How to get data into the most useful structure before transforming, visualizing, and modeling it
• How to clean data and draw plots
• How to manage cognitive resources to support discoveries while exploring data
• The grammar of graphics, literate programming, and how to extract reproducible research that saves time

All these skills provide the foundation that allows data science to happen, and this book shares the best practices for doing things with R to get the results you want and expect.

3. Practical Statistics for Data Scientists

Using R and Python as the foundation, Practical Statistics for Data Scientists, written by Peter Bruce and Andrew Bruce, lays out how statistical methods are essential to data science. It also points out how few courses in data science discuss how vital basic statistics and statistical methods are. In this book, you will learn how to use statistics and avoid misuse of them. Here are some things you will learn in Practical Statistics for Data Scientists:

• The basics of exploratory data analysis and its value as a key preliminary step in data science
• The way the principles of experimental design yields solid answers to questions
• Key classification techniques to predict categories
• How random sampling reduces biases, offering a higher quality data set, even when working with big data

4. Deep Learning: Adaptive Computation and Machine Learning

When you read Deep Learning: Adaptive Computation and Machine Learning, you get to see things from a beginner’s perspective, watching the world of machine learning slowly unfold before you. It covers a broad array of vital topics, including deep-learning techniques and the area’s mathematical and conceptual background. Here are some specific topics the book covers:

• Linear algebra
• Numerical computation
• Machine learning
• Probability theory
• Deep feedforward networks
• Regularization
• Optimization algorithms
• Convolutional networks
• Sequence modeling
• And much more

It also explores potential applications in this area, including:

• Natural language processing
• Speech recognition
• Online recommendation systems
• Bioinformatics
• Computer vision
• Video games

Other invaluable areas covered by this book include numerous theoretical topics like representation learning, structured probabilistic models, Monte Carlo methods, linear factor models, the partition function, deep generative models, and approximate inference.

5. An Introduction to Statistical Learning

An Introduction to Statistical Learning offers a more generalized and less formal and technical treatment of key topics in statistical learning. Winner of the 2014 Eric Ziegel award from Technometrics, translations of the book are available in Chinese, Italian, Japanese, Korean, Mongolian, Russian, and Vietnamese.

Authors Gareth James, Daniela Witten, Trevor Hastie, and Rob Tibshirani have provided a crucial toolkit for data science students, focusing on topics like:

• Bayesian additive regression trees
• Clustering
• Boosting
• Decision trees
• Deep learning
• Multiple testing
• Naive Bayes and generalized linear models
• Sparse methods for classification and regression
• Support vector machines
• Survival analysis

6. Data Science from Scratch

Absolute data science beginners might choose to start here. Reading Data Science from Scratch by Joel Grus offers those discovering data science a place to begin with simple explanations that boost confidence quickly. Here, you will learn about data science libraries, modules, frameworks, and toolkits essential for practicing data science from scratch.

As long as you have an interest in and aptitude for mathematics and some core programming skills, this book was written for you. A few things you will learn in this book:

• The fundamentals of Python with a crash course
• The basics of linear algebra, statistics, and probability, and how each applies to data science
• Gather, analyze, clean, munge, and manipulate data
• Dig into the basics of machine learning
• Use models like Neighbors, Naive Bayes, linear and logistic regression, decision trees, neural networks, and clustering
• Learn more about recommender systems, natural language processing, network analysis, and databases.

7. Naked Statistics

Audie Award Finalist in the Business/Educational category in 2013,Naked Statistics offers a “sexy” look at the field of statistics. Author Charles Wheelan uses everything from baseball batting averages to political polls, game shows, and medical research to provide practical, real-world, and visual applications of statistics.

Wheelan strips away the dry, arcane, and technical details, laying bare the underlying intuition and intrigue that drives statistical analysis. He clarifies crucial concepts like correlation, inference, and regression analysis to provide insight into how careless practices by careless data scientists can manipulate or misrepresent data.

Are You Ready to Apply Data Science Reading to Your Graduate Studies and Beyond?

Are you still wondering: “How can I learn data science?” If so, the best path you can take is enrolling in a graduate data science program focused on helping you discover the meaning in big data. At Lewis University, you will find the information and support you need to learn, grow, and stay competitive in this growing field.

With your extracurricular reading, you’ll continue expanding your knowledge and opportunities throughout your data science studies and career. Earn your master’s degree in data science to learn more about working as a machine learning engineer, data science architect, business intelligence analyst, and much more. To learn more, complete the Request Info, call (815) 836-5610, or reach out to us via email at grad@lewisu.edu.