Book Review: Confident Data Skills
Master the Fundamentals of Working with Data and Supercharge Your Career (Confident Series)
By Kirill Eremenko
Kirill Eremenko is the Founder and CEO of the online educational portal SuperDataScience. I know him from his online course "Machine Learning A-Z™: Hands-On Python & R In Data Science", in which, together with Hadelin de Ponteves, he presents an introduction to the main algorithms and techniques for prediction, classification and clustering models, among others (including neural networks). Due to the good experience I gained with this course (and the SuperDataScience podcasts) I decided to buy his latest book "Confident Data Skills: Master the Fundamentals of Working with Data and Supercharge Your Career (Confident Series)" and then I present what I found most relevant in each chapter and my final opinions of the book.
PART ONE ‘What is it?’ Key Principles
The first part of the book works as a brief introduction to the importance of data science where this field is contextualized in the current world and the mindset that a professional in this area should have is presented.
Kirill takes advantage of Maslow's hierarchy of needs to explain the role of data science in each of the links: physiological needs, security needs, social needs, needs in self-esteem and self-actualization.
Maslow's hierarchy of needs.
The author covers each point in the pyramid and argues its point by presenting real case studies.
This section gave me a better understanding of the role of data science today and why data science has become increasingly popular in the industry in recent years.
PART TWO ‘When and where can I get it?’ Data gathering and analysis
Once the introduction of part one is presented, Kirill develops the rest of the book presenting 'The data science process' in which every stage is meticulously detailed and contains cases from companies such as Netflix and LinkedIn. The whole process is presented without entering into a strong mathematical or statistical background so they are easy for readers of any experience level to understand.
The rest of this part explains the processes of data preparation and analysis. For the preparation part he explains the whole process of data extraction, transformation and loading (ETL). The analysis section explains the most popular machine learning techniques (Classification and clustering) and ends with unsupervised language techniques such as Upper confidence bound and Thompson sampling. All these topics are explained in a simple way without falling into strong mathematical explanations and are easy to understand for all audiences regardless of their level of experience in the subject.
PART THREE ‘How can I present it?’ Communicating data
The third and last part of the book is devoted to the last two parts of the 'Data Science Process'. The aspects that I found most remarkable are the use of color schemes in the visualization of the data. The book ends with tips for building a career in data science, the author discusses topics on how to prepare interviews and the different roles within this field, among other various advices.
Conclusion
Confident Data Skills is a very good book and a hit by Kirill Eremenko. The book has something for every type of user: for the business owner it gives a general idea about the importance of including a data science department and how this branch is of a great help in decision-making. For the user who starts in this field it provides a general idea and allows to identify the path to follow. For the professional in the area, it serves as a book of tips and best practices in the field.