Data Science Without Makeup: A Guidebook for End-Users, Analysts, and Managers

by Mikhail Zhilkin

Overview

Mikhail Zhilkin, a data scientist who has worked on projects ranging from Candy Crush games to Premier League football players’ physical performance, shares his strong views on some of the best and, more importantly, worst practices in data analytics and business intelligence. Why data science is hard, what pitfalls analysts and decision-makers fall into, and what everyone involved can do to give themselves a fighting chance—the book examines these and other questions with the skepticism of someone who has seen the sausage being made.

Honest and direct, full of examples from real life, Data Science Without Makeup: A Guidebook for End-Users, Analysts and Managers will be of great interest to people who aspire to work with data, people who already work with data, and people who work with people who work with data—from students to professional researchers and from early-career to seasoned professionals.

Mikhail Zhilkin is a data scientist at Arsenal FC. He has previously worked on the popular Candy Crush mobile games and in sports betting.

Product Details

ISBN-13:	9780367520687
Publisher:	CRC Press
Publication date:	11/02/2021
Pages:	194
Product dimensions:	5.90(w) x 9.20(h) x 0.60(d)

About the Author

Mikhail Zhilkin is a Data Scientist at Arsenal FC. He has previously worked on the popular Candy Crush mobile games and in sports betting.

Table of Contents

Foreword ix

Preface xiii

Author xv

I The ugly truth

1 What is data science 5

What data science is 5

What data science is for 10

Why it is important to understand your data 14

Where data comes from 16

Glossary 21

Works cited 22

2 Data science is hard 23

Iceberg of details 24

Domino of mistakes 28

No second chance 31

3 Our brain sucks 37

Correlation ≠ causation 38

Reversing Cause and Effect 39

Confounders 39

Outliers 41

Data dredging ("p-hacking") 45

Cognitive biases 50

Confirmation Bias 51

Optimism Bias 53

Information Bias 55

More Work 56

Diluted Argument 56

Lost Purpose 59

Glossary 60

Works cited 60

II A new hope

4 Data science for people 63

Align data science efforts with business needs 64

Mind data science hierarchy of needs 69

Make it simple, reproducible, and shareable 72

Simple 72

Reproducible 74

Shareable 76

Glossary 77

Works cited 77

5 Quality assurance 79

What makes QA difficult? 80

Individual Mindset 80

Team Culture 82

Resources 83

What is there to QA? 84

Data 85

Code 86

Results 88

How to QA all this? 89

Communication 89

Results 92

Code 95

Glossary 96

Works cited 96

6 Automation 97

The automation story 98

Phase 1 "Manual" Data Science 98

Phase 2 Templates 99

Phase 3 Script 100

Phase 4 Full Automation 101

Underappreciated benefits 102

Always Moving Forward 102

Better Quality Assurance 103

Fast Delivery 103

Questions to consider 106

If 106

When 109

How 110

Glossary 117

Works cited 117

III People, people, people

7 Hiring a data scientist 121

Pain 123

Vision 125

Transmission 127

Urgency 132

System 135

P. S. underappreciated qualities 138

Written Communication 139

Goal Orientation 140

Conscientiousness 141

Empathy 141

P. P. S. overappreciated qualities 142

Charisma 142

Confidence 143

Glossary 143

Works cited 144

8 What a data scientist wants 145

Goal 146

Purpose 147

Challenge 148

Achievement 149

Data 149

Autonomy 149

Focus 151

Time 154

Culture 154

Reward 155

Impact 155

Fair Recognition 157

Growth 158

Data scientist types 160

Idea: "Entrepreneur" 162

Theory: "Academic" 162

Tools: "Geek" 162

Solution: "Doer" 163

Recognition: "Careerist" 163

Glossary 165

Works cited 165

9 Measuring performance 167

Time 168

Throughput 169

Goal achievement 171

Opinion 173

Works cited 176

Index 177

Editorial Reviews

Having worked with Mikhail it does not surprise me that he has put together a comprehensive and insightful book on Data Science where down-to-earth pragmatism is the recurring theme. This is a must-read for everyone interested in industrial Data Science, in particular analysts and managers who want to learn from Mikhail‘s great experience and approach.”

—Stefan Freyr Gudmundsson, Lead Data Scientist at H&M, former AI Research Lead at King and Director of Risk Analytics and Modeling at Islandsbanki.

“Mikhail's book is a clear sighted-look at why data science is hard, and why it is so rewarding. It tells the unvarnished truth about data science. The author's view from the trenches will resonate with data scientists, giving them vocabulary and frameworks to describe what they need from colleagues and clients. For sponsors and consumers of data science, this book will clarify what the data scientists are doing with their time. are doing with their time and why they need to do it. Chapter 2 (‘Data Science is Hard’) is worth the price on its own — and then Zhilkin gives us processes to help. An invaluable resource for, in plain language, framing data science as a difficult but valuable role in an organization, with strong advice on processes to maximize the effectiveness of that role. A must-read for any practitioner, manager, or executive sponsor of data science.”

—Ted Lorenzen, Director of Marketing Analytics at Vein Clinics of America

“Mikhail is a pioneer in the applied data science space. His ability to provide innovative solutions to practical questions in a dynamic environment is simply superb. Importantly, Mikhail’s ability to remain calm and composed in high-pressure situations is surpassed only by his humility.”

—Darren Burgess, High Performance Manager at Melbourne FC, former Head of Elite Performance at Arsenal FC

From the Publisher