Book Image

R Data Analysis Cookbook

By : Viswa Viswanathan, Shanthi Viswanathan
Book Image

R Data Analysis Cookbook

By: Viswa Viswanathan, Shanthi Viswanathan

Overview of this book

<p>Data analytics with R has emerged as a very important focus for organizations of all kinds. R enables even those with only an intuitive grasp of the underlying concepts, without a deep mathematical background, to unleash powerful and detailed examinations of their data.</p> <p>This book empowers you by showing you ways to use R to generate professional analysis reports. It provides examples for various important analysis and machine-learning tasks that you can try out with associated and readily available data. The book also teaches you to quickly adapt the example code for your own needs and save yourself the time needed to construct code from scratch.</p>

Related Content you might be interested in

Current Title:

Small book image
R Data Analysis Cookbook
Viswa Viswanathan | Shanthi Viswanathan
May, 2015 | 342 pages
No titles found
Table of Contents (18 chapters)
R Data Analysis Cookbook
R Data Analysis Cookbook
Credits
Credits
About the Authors
About the Authors
About the Reviewers
About the Reviewers
www.PacktPub.com
www.PacktPub.com
Preface
Preface
Free Chapter
1
Acquire and Prepare the Ingredients – Your Data
Acquire and Prepare the Ingredients – Your Data
Introduction
Reading data from CSV files
Reading XML data
Reading JSON data
Reading data from fixed-width formatted files
Reading data from R files and R libraries
Removing cases with missing values
Replacing missing values with the mean
Removing duplicate cases
Rescaling a variable to [0,1]
Normalizing or standardizing data in a data frame
Binning numerical data
Creating dummies for categorical variables
2
What's in There? – Exploratory Data Analysis
What's in There? – Exploratory Data Analysis
Introduction
Creating standard data summaries
Extracting a subset of a dataset
Splitting a dataset
Creating random data partitions
Generating standard plots such as histograms, boxplots, and scatterplots
Generating multiple plots on a grid
Selecting a graphics device
Creating plots with the lattice package
Creating plots with the ggplot2 package
Creating charts that facilitate comparisons
Creating charts that help visualize a possible causality
Creating multivariate plots
3
Where Does It Belong? – Classification
Where Does It Belong? – Classification
Introduction
Generating error/classification-confusion matrices
Generating ROC charts
Building, plotting, and evaluating – classification trees
Using random forest models for classification
Classifying using Support Vector Machine
Classifying using the Naïve Bayes approach
Classifying using the KNN approach
Using neural networks for classification
Classifying using linear discriminant function analysis
Classifying using logistic regression
Using AdaBoost to combine classification tree models
4
Give Me a Number – Regression
Give Me a Number – Regression
Introduction
Computing the root mean squared error
Building KNN models for regression
Performing linear regression
Performing variable selection in linear regression
Building regression trees
Building random forest models for regression
Using neural networks for regression
Performing k-fold cross-validation
Performing leave-one-out-cross-validation to limit overfitting
5
Can You Simplify That? – Data Reduction Techniques
Can You Simplify That? – Data Reduction Techniques
Introduction
Performing cluster analysis using K-means clustering
Performing cluster analysis using hierarchical clustering
Reducing dimensionality with principal component analysis
6
Lessons from History – Time Series Analysis
Lessons from History – Time Series Analysis
Introduction
Creating and examining date objects
Operating on date objects
Performing preliminary analyses on time series data
Using time series objects
Decomposing time series
Filtering time series data
Smoothing and forecasting using the Holt-Winters method
Building an automated ARIMA model
7
It's All About Your Connections – Social Network Analysis
It's All About Your Connections – Social Network Analysis
Introduction
Downloading social network data using public APIs
Creating adjacency matrices and edge lists
Plotting social network data
Computing important network metrics
8
Put Your Best Foot Forward – Document and Present Your Analysis
Put Your Best Foot Forward – Document and Present Your Analysis
Introduction
Generating reports of your data analysis with R Markdown and knitr
Creating interactive web applications with shiny
Creating PDF presentations of your analysis with R Presentation
9
Work Smarter, Not Harder – Efficient and Elegant R Code
Work Smarter, Not Harder – Efficient and Elegant R Code
Introduction
Exploiting vectorized operations
Processing entire rows or columns using the apply function
Applying a function to all elements of a collection with lapply and sapply
Applying functions to subsets of a vector
Using the split-apply-combine strategy with plyr
Slicing, dicing, and combining data with data tables
10
Where in the World? – Geospatial Analysis
Where in the World? – Geospatial Analysis
Introduction
Downloading and plotting a Google map of an area
Overlaying data on the downloaded Google map
Importing ESRI shape files into R
Using the sp package to plot geographic data
Getting maps from the maps package
Creating spatial data frames from regular data frames containing spatial and other data
Creating spatial data frames by combining regular data frames with spatial objects
Adding variables to an existing spatial data frame
11
Playing Nice – Connecting to Other Systems
Playing Nice – Connecting to Other Systems
Introduction
Using Java objects in R
Using JRI to call R functions from Java
Using Rserve to call R functions from Java
Executing R scripts from Java
Using the xlsx package to connect to Excel
Reading data from relational databases – MySQL
Reading data from NoSQL databases – MongoDB
Index
Index

About the Reviewers

Kenneth D. Graves believes that data science will give us superpowers. Or, at the very least, allow us to make better decisions. Toward this end, he has over 15 years of experience in data science and technology, specializing in machine learning, big data, signal processing and marketing, and social media analytics. He has worked for Fortune 500 companies such as CBS and RCA-Technicolor, as well as finance and technology companies, designing state-of-art technologies and data solutions to improve business and organizational decision-making processes and outcomes. His projects have included facial and brand recognition, natural language processing, and predictive analytics. He works and mentors others in many technologies, including R, Python, C++, Hadoop, and SQL.

Kenneth holds degrees and/or certifications in data science, business, film, and classical languages. When he is not trying to discover superpowers, he is a data scientist and acting CTO at Soshag, LLC., a social media analytics firm. He is available for consulting and data science projects throughout the Greater Boston Area. He currently lives in Wellesley, MA.

I wish to thank my wife, Jill, for being the inspiration for all that I do.

Jithin S L completed his BTech in information technology from Loyola Institute of Technology and Science. He started his career in the field of analytics and then moved to various verticals of big data technology. He has worked with reputed organizations, such as Thomson Reuters, IBM, and Flytxt, in different roles. He has worked in the banking, energy, healthcare, and telecom domains, and has handled global projects on big data technology.

He has submitted many research papers on technology and business at national and international conferences. Currently, Jithin is associated with IBM Corporation as a systems analyst—big data big insight in business analytics and optimization unit.

 

“Change is something which brings us to THINK beyond our limits, worries it also provides an opportunity to learn new things in a new way, experiment, explore, and advise towards success.”

 
 --Jithin

I surrender myself to God almighty who helped me review this book in an effective way. I dedicate my work on this book to my dad, Mr. N. Subbian Asari; my lovable mom, Mrs. M. Lekshmi; and my sweet sister, Ms. S.L Jishma, for coordinating and encouraging me to review this book. Last but not least, I would like to thank all my friends.

Dipanjan Sarkar is an IT engineer at Intel, the world's largest silicon company, where he works on analytics and enterprise application development. As part of his experience in the industry so far, he has previously worked as a data engineer at DataWeave, one of India's emerging big data analytics start-ups and also as a graduate technical intern in Intel.

Dipanjan received his master's degree in information technology from the International Institute of Information Technology, Bengaluru. His interests include learning about new technology, disruptive start-ups, and data science. He has also reviewed Learning R for Geospatial Analysis, Packt Publishing.

Hang (Harvey) Yu graduated from the University of Illinois at Urbana-Champaign with a PhD in computational biophysics and a master's degree in statistics. He has extensive experience on data mining, machine learning, and statistics. In the past, Harvey has worked on areas such as stochastic simulations and time series (in C and Python) as part of his academic work. He was intrigued by algorithms and mathematical modeling. He has been involved in data analytics since then.

He is currently working as a data scientist in Silicon Valley. He is passionate about data sciences. He has developed statistical/mathematical models based on techniques such as optimization and predictive modeling in R. Previously, Harvey worked as a computational sciences intern for ExxonMobil.

When Harvey is not programming, he is playing soccer, reading fiction books, or listening to classical music. You can get in touch with him at or on LinkedIn at www.linkedin.com/in/hangyu1.

Previous Section
Next Chapter