The One That You Need: Data Science Book Recommendations
So many resources, so little time.
This problem happens a lot to those who want to learn something new (or maybe — not new). Juggling between papers, workloads, and other personal stuff while expecting ourselves to be ‘more productive’ by learning could be an infinite guilt-tripping experience. Moreover, if the ‘new’ thing that you want to learn is data science, it would feel like a double guilt-trip experience because it’s hard, but we want to master it, but time is limited, yet we want to improve.
It’s endless.
How if I told you that you just need ONE book which you’ll need to stick with until the end, then you’ll be learning a lot more than you can imagine?
So, what is THE ONE book that YOU need? Refer yourself to the flowchart below, then find which group is suitable to your current skillsets.
FYI, why do the ‘basic Python’ and ‘basic stat & linear algebra’ become the factors in deciding which group you belong to? It’s because data science is essentially a blend of coding skills and advanced math & statistics. Oh, and also domain knowledge according to your field, such as economics, social science, business management, etc. However, now we can focus on both coding skills & math/stats first.
Let’s jump in right away, shall we?
- Group 1: Data Science From Scratch by Joel Grus
This book gives you all-round materials (both basic coding & math/stats) in one book! Based on the book description, it covers:
- A crash course in Python,
- The basics of linear algebra, statistics, and probability — and understand how and when they’re used in data science,
- Implement models such as k-nearest neighbors, Naive Bayes, linear and logistic regression, decision trees, neural networks, and clustering.
Perfect for beginners!
2. Group 2: Python Data Science Handbook by Jake VanderPlas
This book is great for those who have learned statistics & math but haven’t learned Python before. It covers:
- IPython,
- Numpy, pandas, matplotlib,
- Scikit-learn.
This book will help you complement your knowledge in math and stats with Python coding skills. Once you’re done with this book, you can practice it right away!
3. Group 3: Think Stats by Allen B. Downey
The ONE that all programmers need when they want to learn stats for data science! This book covers the basic stats needed, such as:
- Statistical thinking,
- Descriptive statistics,
- Hypothesis testing, etc.
However, if you need a deeper understanding of machine learning modeling in stats & math point-of-view, you can refer to MML (Math for Machine Learning). This book is written by Marc Peter Deisenroth, A. Aldo Faisal, and Cheng Soon Ong. I think this book is free! Shout out to the writers!
4. Group 4: The Data Science Handbook by Carl Shan
Okay, you’ve mastered Python coding and statistics (plus machine learning). Now, what? Please read this book. This book is the only one out there (as far as I know) that summarize all in-depth interviews with 25 world’s best data scientist! Spoiler: they are from Facebook, LinkedIn, Pandora, Intuit, The New York Times, Airbnb, Uber, etc!
They talk about their careers, personal stories, perspectives on data science, and life advice, which is very much needed to get more inspiration and a spirit-boost for anyone in the data science path.
That’s it?
Yes, that’s it. You can explore more, but please be aware that data science heavily relies on practical skills. You need to practice, practice, and lots of practice. It’s a never-ending learning process since data science is evolving rapidly throughout the years! Definitely, a fun learning experience, isn’t it? :)
