Eremenko - Confident Data Skills

To my parents, Alexander and Elena Eremenko, who taught me the most important thing in life how to be a good person

Thank you for picking up this book. Youve made a huge step in your journey into data science.

All readers gain complimentary access to my Data Science AZ course.

Just go to www.superdatascience.com/bookbonus and use the password datarockstar.

You can download a guide to using colour in visualizations at www.superdatascience.com/cds .

Happy analysing!

I would like to thank my father, Alexander Eremenko, whose love and care have shaped me into the person I am today, and who has shown me through his firm guidance how I can take hold of lifes opportunities. Thank you to my warm-hearted mother, Elena Eremenko, for always lending an ear to my crazy ideas and for encouraging my brothers and me to take part in the wider world through music, language, dance and so much more. Were it not for her wise counsel, I would never have immigrated to Australia.

Thanks to my brother, Mark Eremenko, for always believing in me, and for his unshakeable confidence. His fearlessness in life continues to fuel so many of the important decisions I make. Thank you to my brother, Ilya Eremenko, wise beyond his years, for his impressive business ideas and well-considered ventures. I am certain that fame and fortune will soon be knocking on his door.

Thank you to my grandmother Valentina, aunt Natasha and cousin Yura, for their endless love and care. And thank you to the entire Tanakovic and Svoren families, including my brothers Adam and David for all the dearest moments we share.

Thank you to my students and to the thousands of people who listen to the SuperDataScience podcast. My audience inspires me to keep going!

Several key people helped to make this book a reality. I would like to thank my writing partner Zara Karschay for capturing my voice. Thank you to my commissioning editor Rebecca Bush and production editor Stefan Leszczuk, along with Anna Moss, whose feedback and guidance were fundamental to the writing process, and to Kogan Pages editorial team for their rigour. Thank you to my friend and business partner Hadelin de Ponteves for inspiring me and for being a great source of support in tackling some of the most challenging subject matters in the field of data science, as well as for his help in reviewing the technical aspects of this book.

Thank you to my friend and executive assistant Mitja Bosni for his tireless efforts in making this second edition possible. My thanks also go to the talented team at SuperDataScience for taking on additional responsibilities so that I could write this book. Thank you to the hardworking team at Udemy, including my supportive account managers Lana Martinez and Erin Adams.

Thank you to my friend and mentor Artem Vladimirov, whose admirable work ethic and knowledge lay the foundations upon which I have built everything I know about data science. Many thanks to Vitaly Dolgov, Ivor Lok, Richard Hopkins, Tracy Crossley and Herb Kanis for being excellent role models, for believing in me, for always being there when I needed help, and for guiding me through times both good and bad. Thank you to Katherina Andryskova I am grateful that she was the first person to read and offer invaluable feedback on Confident Data Skills.

I give my express thanks to the people whose work contributed to the case studies in this book: Alberto Cairo, Samuel Hinton, Richard Hopkins, Kristen Kehrer, Raul Popa, Caroline McColl, Ulf Morys, Daniel and Leigh Pullen, Dominic Roe, Adrian Rosebrock, Matthew Rosenquist, Dan Shiebler, Ben Taylor, Artem Vladimirov, and Stephen Welch.

The motivational teachers at Moscow School 54, the Moscow Institute of Physics and Technology and the University of Queensland have my thanks for giving me such beneficial education. To all my past colleagues at Deloitte and Sunsuper, thank you for giving me the professional development I needed for creating my data science toolkit.

Most of all, I want to thank you, the reader, for giving me your valuable time. It has been my foremost hope that this book will encourage those who wish to understand and implement data science in their careers.

A great deal has changed in the two years since the first edition. The field of data science has stormed ahead everything is bigger, faster and stronger. Increased processing capabilities have brought reinforcement learning and deep learning to the market. Neural networks are much easier to build. Artificial intelligence has changed the game in many industries. But we must also not forget about the dark side of this moon. Ruthless hacks have held entire city systems hostage. Many fear the prospect of machines gaining sentience, and of data-driven companies abusing our privacy.

I might be a small cog in the data science mechanism, but I have learned that anyone has the potential to bring about change. I continue to encourage my students to be a part of the movement, and I practise what I preach: my 2018 co-founded corporate training start-up BlueLife AI works from the idea that everyone can benefit from the power of artificial intelligence.

Nevertheless, I cant take praise for this move without acknowledging my colleague and course co-developer, Hadelin de Ponteves, and my students. My and Hadelins online courses have become a vibrant community of over a million data scientists from more than 200 countries, all of whom are eager to share and critique ideas. The international DataScienceGO conference, now entering its fourth iteration, doubles its delegate numbers every year. As the conference welcomes all abilities, we originally devised pathways to accommodate different skill sets. Yet we learned the most exciting panels were those that bridged multiple routes. In the sessions where people of diverse capabilities interacted with each other, ideas spread. It became clear that data scientists are friendly, engaged people who want to learn in an open and welcoming environment that brings everyone together.

I have learned, above all, that human networks are powerful. That data scientists love a sense of community. That our students want to continue learning with us, even when theyve surpassed the courses we have available. We and I mean that generally get so much value from helping each other. I once thought that having a reliable network of peers was just one of many components to success in the field. Now I see that it is the most critical element of all.

We can make the same argument for data. A single data point rarely makes a difference, but drop it into a network of a million others, and it can contribute to real change. Here is my advice: join a network of data scientists. Follow the movers and shakers on LinkedIn and Twitter. Take a course and talk to your fellow students. Get involved in the discussion. Youll learn from others, and gain perspective of the latest developments, and in time will gain the confidence to make valuable contributions of your own.

With all the attention given to the apparently limitless potential of technology and the extensive opportunities for keen entrepreneurs, some may ask why they should bother with data science at all why not simply learn the principles of technology? After all, technology powers the world, and it shows no signs of slowing down. Any reader with an eye to their career might think that learning how to develop new technologies would surely be the better way forward.