Statistics, Data Mining, and Machine Learning in Astronomy: A Practical Python Guide for the Analysis of Survey Data, Updated Edition

by Željko Ivezić, Andrew J. Connolly, Jacob T. VanderPlas

Statistics, Data Mining, and Machine Learning in Astronomy is the essential introduction to the statistical methods needed to analyze complex data sets from astronomical surveys such as the Panoramic Survey Telescope and Rapid Response System, the Dark Energy Survey, and the Large Synoptic Survey Telescope. Now fully updated, it presents a wealth of practical analysis problems, evaluates the techniques for solving them, and explains how to use various approaches for different types and sizes of data sets. Python code and sample data sets are provided for all applications described in the book. The supporting data sets have been carefully selected from contemporary astronomical surveys and are easy to download and use. The accompanying Python code is publicly available, well documented, and follows uniform coding standards. Together, the data sets and code enable readers to reproduce all the figures and examples, engage with the different methods, and adapt them to their own fields of interest.



An accessible textbook for students and an indispensable reference for researchers, this updated edition features new sections on deep learning methods, hierarchical Bayes modeling, and approximate Bayesian computation. The chapters have been revised throughout and the astroML code has been brought completely up to date.
Fully revised and expanded


Describes the most useful statistical and data-mining methods for extracting knowledge from huge and complex astronomical data sets


Features real-world data sets from astronomical surveys


Uses a freely available Python codebase throughout


Ideal for graduate students, advanced undergraduates, and working astronomers

Genres: Astronomy, Reference, Mathematics, Coding, Science, Textbooks, Computer Science

552 pages, Kindle Edition

First published May 4, 2013

Reviews

[gully · Rating 5 out of 5]

Excellent book. The book comes with a companion website. Here you can find the source code for every single textbook figure. How cool is that! There is an active GitHub repo for all the textbook errata too. Highly recommended for 1) All astronomy graduate students and postdocs, 2) anyone interested in data intensive applications, 3) Professors teaching a modern astronomy methods course.
This is simply a must have book for 2015.

[Ravi · Rating 4 out of 5]

Pretty good balance of depth and breadth. Having the AstroML code be open source is great. I wish there was more on neural networks (e.g., CNNs, GANs), which are mentioned in the book as if they were explained, but actually are not touched upon. Also there is a lot of mention of cross validation before the term is ever defined. Overall a very solid stats and ML reference.

[Benji · Rating 4 out of 5]

Practical indeed, contains intuitive explanations of (astro)statistical analysis and data mining techniques while presenting a wealth of tips about trade-offs in practice.