Python Data Science Essentials, Second Edition

Alberto Boschetti and Luca Massaron

Published by
Packt Publishing
ISBN 9781786462138
RRP £30.99
Reviewed by Prof. Kalum Priyanath Udagepola
Score

9 out of 10

If someone contextualizes their practical challenges as guidelines, it is a great upper hand for the learners. Alberto Boschetti and Luca Massaron give advice with clearly set out boundaries to contextualisation to ensure readers can readily determine what is acceptable to the industry. This advice develops around scenarios, examples and codes of data science projects.

The authors are data scientists with expertise in statistics, linking with other sophisticated technical subject fields. This book has simplified the complexities that are relevant to beginners and intermediate data scientists with their understanding may have faced in using Python. In this book, users are recommended Python 3.4 or above for all its examples to practice.

The book engages and absorbs the reader into the subject matter involving almost all the human senses. The beauty of the book is that it has six chapters linked with resources (data and source codes). These resources are of immense value and will surely intrigue both beginners, and intermediate users. At the beginning of each chapter readers are able to clearly visualise what will be learnt during the chapter. The book gives more extensive knowledge about practical data mining principals through scientific methodology and effectively tests the performance of the user's machine learning hypothesis.

If the reader studies the book and completes the lab practice, it is a great chance to enhance user data manipulation and machine learning skills.

In this second edition, it is evident the authors have invested both time and effort, and have listened to user feedback to improve this particular edition. This edition displays more maturity and delivers more focus on updated and expanded content. Chapter four on Machine Learning in this second edition is an excellent move I think, as it’s one of the most widely used data science techniques with python.

Visualize the machine learning and optimisation processes the authors discuss in chapter 3, ‘The Data Pipeline’, and chapter 4, ‘Machine Learning’. If readers choose to get colour images of this book, there is the facility, and I am sure it is a bonus for the readers.

I recommend this book to all data science labs if they are dedicated to investing real industry experiences to successfully obtain their future research project deliverables.

This 354 page book is an excellent guide on learning data science through python for those aspiring to become experienced in it. It is also one of the few books that one will find truly practical and engaging.

Further information: Packt

September 2017