Modern Data Science with R

Contents
Tables
Figures
Preface
Intro to Data Science
	1 Prologue - why data science?
		1.1 What is data science?
		1.2 Case study: The evolution of sabermetrics
		1.3 Datasets
		1.4 Further resources
	2 Data visualization
		2.1 The 2012 federal election cycle
			2.1.1 Are these two groups different?
			2.1.2 Graphing variation
			2.1.3 Examining relationships among variables
			2.1.4 Networks
		2.2 Composing data graphics
			2.2.1 A taxonomy for data graphics
			2.2.2 Color
			2.2.3 Dissecting data graphics
		2.3 Importance of data graphics: Challenger
		2.4 Creating effective presentations
		2.5 The wider world of data visualization
		2.6 Further resources
		2.7 Exercises
	3 Grammar for graphics
		3.1 A grammar for data graphics
			3.1.1 Aesthetics
			3.1.2 Scale
			3.1.3 Guides
			3.1.4 Facets
			3.1.5 Layers
		3.2 Canonical data graphics in R
			3.2.1 Univariate displays
			3.2.2 Multivariate displays
			3.2.3 Maps
			3.2.4 Networks
		3.3 Extended example: Historical baby names
			3.3.1 Percentage of people alive today
			3.3.2 Most common women's names
		3.4 Further resources
		3.5 Exercises
	4 Data wrangling
		4.1 A grammar for data wrangling
			4.1.1 select() and filter()
			4.1.2 mutate() and rename()
			4.1.3 arrange()
			4.1.4 summarize() with group_by()
		4.2 Extended example: Ben's time with the Mets
		4.3 Combining multiple tables
			4.3.1 inner_join()
			4.3.2 left_join()
		4.4 Extended example: Manny Ramirez
		4.5 Further resources
		4.6 Exercises
	5 Tidy data & iteration
		5.1 Tidy data
			5.1.1 Motivation
			5.1.2 What are tidy data?
		5.2 Reshaping data
			5.2.1 Data verbs for converting wide to narrow and vice versa
			5.2.2 Spreading
			5.2.3 Gathering
			5.2.4 Example: Gender-neutral names
		5.3 Naming conventions
		5.4 Automation and iteration
			5.4.1 Vectorized operations
			5.4.2 The apply() family of functions
			5.4.3 Iteration over subgroups with dplyr: :do()
			5.4.4 Iteration with mosaic: :do
		5.5 Data intake
			5.5.1 Data-table friendly formats
			5.5.2 APIs
			5.5.3 Cleaning data
			5.5.4 Example: Japanese nuclear reactors
		5.6 Further resources
		5.7 Exercises
	6 Professional Ethics
		6.1 Introduction
		6.2 Truthful falsehoods
		6.3 Some settings for professional ethics
			6.3.1 The chief executive officer
			6.3.2 Employment discrimination
			6.3.3 Data scraping
			6.3.4 Reproducible spreadsheet analysis
			6.3.5 Drug dangers
			6.3.6 Legal negotiations
		6.4 Some principles to guide ethical action
			6.4.1 Applying the precepts
		6.5 Data and disclosure
			6.5.1 Reidentification and disclosure avoidance
			6.5.2 Safe data storage
			6.5.3 Data scraping and terms of use
		6.6 Reproducibility
			6.6.1 Example: Erroneous data merging
		6.7 Professional guidelines for ethical conduct
		6.8 Ethics, collectively
		6.9 Further resources
		6.10 Exercises
Statistics & Modeling
	7 Statistical Foundations
		7.1 Samples and populations
		7.2 Sample statistics
		7.3 The bootstrap
		7.4 Outliers
		7.5 Statistical models: Explaining variation
		7.6 Confounding and accounting for other factors
		7.7 The perils of p-values
		7.8 Further resources
		7.9 Exercises
	8 Statistical Learning & Predictive Analytics
		8.1 Supervised learning
		8.2 Classifiers
			8.2.1 Decision trees
			8.2.2 Example: High-earners in the 1994 United States Census
			8.2.3 Tuning parameters
			8.2.4 Random forests
			8.2.5 Nearest neighbor
			8.2.6 Naïve Bayes
			8.2.7 Artificial neural networks
		8.3 Ensemble methods
		8.4 Evaluating models
			8.4.1 Cross-validation
			8.4.2 Measuring prediction error
			8.4.3 Confusion matrix
			8.4.4 ROC curves
			8.4.5 Bias-variance trade-off
			8.4.6 Example: Evaluation of income models
		8.5 Extended example: Who has diabetes?
		8.6 Regularization
		8.7 Further resources
		8.8 Exercises
	9 Unsupervised Learning
		9.1 Clustering
			9.1.1 Hierarchical clustering
			9.1.2 k-means
		9.2 Dimension reduction
			9.2.1 Intuitive approaches
			9.2.2 Singular value decomposition
		9.3 Further resources
		9.4 Exercises
	10 Simulation
		10.1 Reasoning in reverse
		10.2 Extended example: Grouping cancers
		10.3 Randomizing functions
		10.4 Simulating variability
			10.4.1 The partially planned rendezvous
			10.4.2 The jobs report
			10.4.3 Restaurant health and sanitation grades
		10.5 Simulating a complex system
		10.6 Random networks
		10.7 Key principles of simulation
		10.8 Further resources
		10.9 Exercises
Topics in Data Science
	11 Interactive data graphics
		11.1 Rich Web content using D3. js and htmlwidgets
			11.1.1 Leaet
			11.1.2 Plot.ly
			11.1.3 DataTables
			11.1.4 dygraphs
			11.1.5 streamgraphs
		11.2 Dynamic visualization using ggvis
		11.3 Interactive Web apps with Shiny
		11.4 Further customization
		11.5 Extended example: Hot dog eating
		11.6 Further resources
		11.7 Exercises
	12 Database querying using SQL
		12.1 From dplyr to SQL
		12.2 Flat-file databases
		12.3 The SQL universe
		12.4 The SQL data manipulation language
			12.4.1 SELECT...FROM
			12.4.2 WHERE
			12.4.3 GROUP BY
			12.4.4 ORDER BY
			12.4.5 HAVING
			12.4.6 LIMIT
			12.4.7 JOIN
			12.4.8 UNION
			12.4.9 Subqueries
		12.5 Extended example: FiveThirtyEight flights
		12.6 SQL vs. R
		12.7 Further resources
		12.8 Exercises
	13 Database administration
		13.1 Constructing efficient SQL databases
			13.1.1 Creating new databases
			13.1.2 CREATE TABLE
			13.1.3 Keys
			13.1.4 Indices
			13.1.5 EXPLAIN
			13.1.6 Partitioning
		13.2 Changing SQL data
			13.2.1 UPDATE
			13.2.2 INSERT
			13.2.3 LOAD DATA
		13.3 Extended example: Building a database
			13.3.1 Extract
			13.3.2 Transform
			13.3.3 Load into MySQL database
		13.4 Scalability
		13.5 Further resources
		13.6 Exercises
	14 Working with spatial data
		14.1 Motivation: What's so great about spatial data?
		14.2 Spatial data structures
		14.3 Making maps
			14.3.1 Static maps with ggmap
			14.3.2 Projections
			14.3.3 Geocoding, routes, and distances
			14.3.4 Dynamic maps with leaflet
		14.4 Extended example: Congressional districts
			14.4.1 Election results
			14.4.2 Congressional districts
			14.4.3 Putting it all together
			14.4.4 Using ggmap
			14.4.5 Using leaflet
		14.5 Effective maps: How (not) to lie
		14.6 Extended example: Historical airline route maps
			14.6.1 Using ggmap
			14.6.2 Using leaflet
		14.7 Projecting polygons
		14.8 Playing well with others
		14.9 Further resources
		14.10 Exercises
	15 Text as data
		15.1 Tools for working with text
			15.1.1 Regular expressions using Macbeth
			15.1.2 Example: Life and death in Macbeth
		15.2 Analyzing textual data
			15.2.1 Corpora
			15.2.2 Word clouds
			15.2.3 Document term matrices
		15.3 Ingesting text
			15.3.1 Example: Scraping the songs of the Beatles
			15.3.2 Scraping data from Twitter
		15.4 Further resources
		15.5 Exercises
	16 Network science
		16.1 Introduction to network science
			16.1.1 Definitions
			16.1.2 A brief history of network science
		16.2 Extended example: Six degrees of Kristen Stewart
			16.2.1 Collecting Hollywood data
			16.2.2 Building the Hollywood network
			16.2.3 Building a Kristen Stewart oracle
		16.3 PageRank
		16.4 Extended example: 1996 men's college basketball
		16.5 Further resources
		16.6 Exercises
	17 Epilogue - towards "big data"
		17.1 Notions of big data
		17.2 Tools for bigger data
			17.2.1 Data and memory structures for big data
			17.2.2 Compilation
			17.2.3 Parallel and distributed computing
			17.2.4 Alternatives to SQL
		17.3 Alternatives to R
		17.4 Closing thoughts
		17.5 Further resources
Packages used in this book
	A.1 The mdsr package
	A.2 The etl package suite
	A.3 Other packages
	A.4 Further resources
Intro to R & RStudio
	B.1 Installation
		B.1.1 Installation under Windows
		B.1.2 Installation under Mac OS X
		B.1.3 Installation under Linux
		B.1.4 RStudio
	B.2 Running RStudio and sample session
	B.3 Learning R
		B.3.1 Getting help
		B.3.2 swirl
	B.4 Fundamental structures and objects
		B.4.1 Objects and vectors
		B.4.2 Operators
		B.4.3 Lists
		B.4.4 Matrices
		B.4.5 Dataframes
		B.4.6 Attributes and classes
		B.4.7 Options
		B.4.8 Functions
	B.5 Add-ons: Packages
		B.5.1 Introduction to packages
		B.5.2 CRAN task views
		B.5.3 Session information
		B.5.4 Packages and name conflicts
		B.5.5 Maintaining packages
		B.5.6 Installed libraries and packages
	B.6 Further resources
	B.7 Exercises
Algorithmic thinking
	C.1 Introduction
	C.2 Simple example
	C.3 Extended example: Law of large numbers
	C.4 Non-standard evaluation
	C.5 Debugging and defensive coding
	C.6 Further resources
	C.7 Exercises
Reproducible analysis & workflow
	D.1 Scriptable statistical computing
	D.2 Reproducible analysis with R Markdown
	D.3 Projects and version control
	D.4 Further resources
	D.5 Exercises
Regression modeling
	E.1 Simple linear regression
		E.1.1 Motivating example: Modeling usage of a rail trail
		E.1.2 Model visualization
		E.1.3 Measuring the strength of fit
		E.1.4 Categorical explanatory variables
	E.2 Multiple regression
		E.2.1 Parallel slopes: Multiple regression with a categorical variable
		E.2.2 Parallel planes: Multiple regression with a second quantitative variable
		E.2.3 Non-parallel slopes: Multiple regression with interaction
		E.2.4 Modelling non-linear relationships
	E.3 Inference for regression
	E.4 Assumptions underlying regression
	E.5 Logistic regression
	E.6 Further resources
	E.7 Exercises
Setting up a database server
	F.1 SQLite
	F.2 MySQL
		F.2.1 Installation
		F.2.2 Access
		F.2.3 Running scripts from the command line
	F.3 PostgreSQL
	F.4 Connecting to SQL
		F.4.1 The command line client
		F.4.2 GUIs
		F.4.3 R and RStudio
		F.4.4 Load into SQLite database
Biblio
Index
R index