Becoming a Data Head: How to Think, Speak, and Understand Data Science, Statistics, and Machine Learning

Cover
Title Page
Copyright Page
About the Authors
About the Technical Editors
Acknowledgments
Contents
Introduction
	The Data Science Industrial Complex
	Why We Care
	Data in the Workplace
	You Can Understand the Big Picture
	Who This Book Is Written For
	Why We Wrote This Book
	What You’ll Learn
	How This Book Is Organized
	One Last Thing Before We Begin
Part I Thinking Like a Data Head
	Chapter 1 What Is the Problem?
		Questions a Data Head Should Ask
			Why Is This Problem Important?
			Who Does This Problem Affect?
			What If We Don’t Have the Right Data?
			When Is the Project Over?
			What If We Don’t Like the Results?
		Understanding Why Data Projects Fail
			Customer Perception
			Discussion
		Working on Problems That Matter
		Chapter Summary
	Chapter 2 What Is Data?
		Data vs. Information
			An Example Dataset
		Data Types
		How Data Is Collected and Structured
			Observational vs. Experimental Data
			Structured vs. Unstructured Data
		Basic Summary Statistics
		Chapter Summary
	Chapter 3 Prepare to Think Statistically
		Ask Questions
		There Is Variation in All Things
			Scenario: Customer Perception (The Sequel)
			Case Study: Kidney-Cancer Rates
		Probabilities and Statistics
			Probability vs. Intuition
			Discovery with Statistics
		Chapter Summary
Part II Speaking Like a Data Head
	Chapter 4 Argue with the Data
		What Would You Do?
			Missing Data Disaster
		Tell Me the Data Origin Story
			Who Collected the Data?
			How Was the Data Collected?
		Is the Data Representative?
			Is There Sampling Bias?
			What Did You Do with Outliers?
		What Data Am I Not Seeing?
			How Did You Deal with Missing Values?
			Can the Data Measure What You Want It to Measure?
		Argue with Data of All Sizes
		Chapter Summary
	Chapter 5 Explore the Data
		Exploratory Data Analysis and You
		Embracing the Exploratory Mindset
			Questions to Guide You
			The Setup
		Can the Data Answer the Question?
			Set Expectations and Use Common Sense
			Do the Values Make Intuitive Sense?
			Watch Out: Outliers and Missing Values
		Did You Discover Any Relationships?
			Understanding Correlation
			Watch Out: Misinterpreting Correlation
			Watch Out: Correlation Does Not Imply Causation
		Did You Find New Opportunities in the Data?
		Chapter Summary
	Chapter 6 Examine the Probabilities
		Take a Guess
		The Rules of the Game
			Notation
			Conditional Probability and Independent Events
			The Probability of Multiple Events
		Probability Thought Exercise
			Next Steps
		Be Careful Assuming Independence
			Don’t Fall for the Gambler’s Fallacy
		All Probabilities Are Conditional
			Don’t Swap Dependencies
			Bayes’ Theorem
		Ensure the Probabilities Have Meaning
			Calibration
			Rare Events Can, and Do, Happen
		Chapter Summary
	Chapter 7 Challenge the Statistics
		Quick Lessons on Inference
			Give Yourself Some Wiggle Room
			More Data, More Evidence
			Challenge the Status Quo
			Evidence to the Contrary
			Balance Decision Errors
		The Process of Statistical Inference
		The Questions You Should Ask to Challenge the Statistics
			What Is the Context for These Statistics?
			What Is the Sample Size?
			What Are You Testing?
			What Is the Null Hypothesis?
			What Is the Significance Level?
			How Many Tests Are You Doing?
			Can I See the Confidence Intervals?
			Is This Practically Significant?
			Are You Assuming Causality?
		Chapter Summary
Part III Understanding the Data Scientist’s Toolbox
	Chapter 8 Search for Hidden Groups
		Unsupervised Learning
		Dimensionality Reduction
			Creating Composite Features
		Principal Component Analysis
			Principal Components in Athletic Ability
			PCA Summary
			Potential Traps
		Clustering
		k-Means Clustering
			Clustering Retail Locations
			Potential Traps
		Chapter Summary
	Chapter 9 Understand the Regression Model
		Supervised Learning
		Linear Regression: What It Does
			Least Squares Regression: Not Just a Clever Name
		Linear Regression: What It Gives You
			Extending to Many Features
		Linear Regression: What Confusion It Causes
			Omitted Variables
			Multicollinearity
			Data Leakage
			Extrapolation Failures
			Many Relationships Aren’t Linear
			Are You Explaining or Predicting?
			Regression Performance
		Other Regression Models
		Chapter Summary
	Chapter 10 Understand the Classification Model
		Introduction to Classification
			What You’ll Learn
			Classification Problem Setup
		Logistic Regression
			Logistic Regression: So What?
		Decision Trees
		Ensemble Methods
			Random Forests
			Gradient Boosted Trees
			Interpretability of Ensemble Models
		Watch Out for Pitfalls
			Misapplication of the Problem
			Data Leakage
			Not Splitting Your Data
			Choosing the Right Decision Threshold
		Misunderstanding Accuracy
			Confusion Matrices
		Chapter Summary
	Chapter 11 Understand Text Analytics
		Expectations of Text Analytics
		How Text Becomes Numbers
			A Big Bag of Words
			N-Grams
			Word Embeddings
		Topic Modeling
		Text Classification
			Naïve Bayes
			Sentiment Analysis
		Practical Considerations When Working with Text
			Big Tech Has the Upper Hand
		Chapter Summary
	Chapter 12 Conceptualize Deep Learning
		Neural Networks
			How Are Neural Networks Like the Brain?
			A Simple Neural Network
			How a Neural Network Learns
			A Slightly More Complex Neural Network
		Applications of Deep Learning
			The Benefits of Deep Learning
			How Computers “See” Images
			Convolutional Neural Networks
			Deep Learning on Language and Sequences
		Deep Learning in Practice
			Do You Have Data?
			Is Your Data Structured?
			What Will the Network Look Like?
		Artificial Intelligence and You
			Big Tech Has the Upper Hand
			Ethics in Deep Learning
		Chapter Summary
Part IV Ensuring Success
	Chapter 13 Watch Out for Pitfalls
		Biases and Weird Phenomena in Data
			Survivorship Bias
			Regression to the Mean
			Simpson’s Paradox
			Confirmation Bias
			Effort Bias (aka the “Sunk Cost Fallacy”)
			Algorithmic Bias
			Uncategorized Bias
		The Big List of Pitfalls
			Statistical and Machine Learning Pitfalls
			Project Pitfalls
		Chapter Summary
	Chapter 14 Know the People and Personalities
		Seven Scenes of Communication Breakdowns
			The Postmortem
			Storytime
			The Telephone Game
			Into the Weeds
			The Reality Check
			The Takeover
			The Blowhard
		Data Personalities
			Data Enthusiasts
			Data Cynics
			Data Heads
		Chapter Summary
	Chapter 15 What’s Next?
Index
EULA