Recommended Books

These are the best books that I've read in terms of helping me become a better data scientist. Purchasing books from these links helps support Data Science Bytes.

Data Science and Machine Learning

• Data Science for Business A great description of many machine learning techniques with discussions of their application to real problems.

• Python for Data Analysis A thorough and well written description of the python data science stack by one of the people who made it what it is.

• Building Machine Learning Systems with Python This book goes in depth into the process of building machine learning systems, including feature extraction, evaluating the results and iterating on them to improve performance. Unfortunately the code available online as a supplement for this book doesn't match well with the code in the book and doesn't always work without modification.

• Effective Java Effective Java is a masterpiece and a classic in the field of computer science. Even if you don't use java you'll write better code after reading this book, which makes your data science better.

• Think Bayes A good introduction to the Bayesian thinking and methods using python code. I'd recommend this book to any scientist. I also consider it a good primer on using classes in python. PDF and HTML versions are available for free at Green Tea Press.

• Think Stats I especially recommend this book if you're coming into data science from a non-statistics background. It provides a useful (and fast) introduction to many important statistical concepts using python code. PDF and HTML versions are available for free at Green Tea Press.

• Natural Language Processing with Python An introduction to dealing with natural language data using the nltk library for python. The book starts out with basic python programming concepts and moves into dealing with natural language data and language processing tasks, including some machine learning techniques. An html version is available at the nltk website.

Preparing for Data Science Interviews

Unless you're coming from a computer science or, depending on the company, statistics, background the problems you encounter during a data science interview are likely different from the problems you've encountered in the past. These books can help you prepare for the types of questions typical of interviews at technology companies.

• Cracking the Coding Interview: 150 Programming Questions and Solutions This book contains a good mix of coding questions including many focusing on combinatorics and data structures, which you'll likely encounter at a data science interview. There is also advice about interviewing, information about typical interview structure and other helpful tips. The solution for the code are in Java, however I tried solving many of the problems in python.

• Think Complexity Definitely very useful for interviews as graph search questions come up frequently. This book provides a well paced introduction to several interesting algorithms and phenomena of computer science and is definitely worth a read. PDF and HTML versions are available for free at Green Tea Press.