IA Crash Course in Data Science
1.What is Data Science?
- Moneyball
- Voter Turnout
- Engineering Solutions
2.What is Statistics Good For?
3.What is Machine Learning?
4.What is Software Engineering for Data Science?
5.Structure of a Data Science Project
6.Output of a Data Science Experiment
7.Defining Success: Four Secrets of a Successful Data Science Experiment
8.Data Science Toolbox
9.Separating Hype from Value
- IIBuilding the Data Science Team
10.The Data Team
11.When Do You Need Data Science?
- The Startup
- The Mid-Sized Organization
- Large Organizations
12.Qualifications & Skills
- Data Engineer
- Data Scientist
- Data Science Manager
13.Assembling the Team
- Where to Find the Data Team
- Interviewing for Data Science
14.Management Strategies
- Onboarding the Data Science Team
- Managing the Team
15.Working With Other Teams
- Embedded vs. Dedicated
- How Does Data Science Interact with Other Groups?
- Empowering Others to Use Data
16.Common Difficulties
- Interaction Difficulties
- Internal Difficulties
- IIIManaging Data Analysis
17.The Data Analysis Iteration
- Epicycle of Analysis
18.Asking the Question
- Types of Questions
19.Exploratory Data Analysis
20.Modeling
- What Are the Goals of Formal Modeling?
- Associational Analyses
- Prediction Analyses
21.Interpretation
22.Communication
- IVData Science in Real Life
23.What You’ve Gotten Yourself Into
- Data double duty
- Multiplicity
- Randomization versus observational studies
24.The Data Pull is Clean
25.The Experiment is Carefully Designed: Principles
- Causality
- Confounding
26.The Experiment is Carefully Designed: Things to Do
- A/B testing
- Sampling
- Blocking and Adjustment
27.Results of the Analysis Are Clear
- Multiple comparisons
- Effect sizes, significance, modeling
- Comparison with benchmark effects
- Negative controls
28.The Decision is Obvious
- The decision is (not) obvious
- Estimation target is relevant
29.Analysis Product is Awesome
About the Authors