Introduction to the RevoScaleR package that ships with Revolution R Enterprise. We discuss the challenges associated with dealing with Big Data and how the functions and algorithms in RevoScaleR address them.

Big Data Analytics Overview

Challenges with Big Data

Demonstration of Revolution R Enterprise for Big Data Analytics

File Format

Importing Data with rxImport Function

Functions for Summarizing Data

Creating New Variables using rxDataStep

Transforming Variables Using rxDataStep

Correlations

Linear Regression

How Parallel Exeternal Memory Algorithms in RevoScaleR work

How the Algorithms in RevoScaleR work

Introduction

Chapter 1

Intro to the RevoScaleR functions used to explore large datasets. Learn how to use these functions to summarize, cross-tabulate, and visualize variables in large datasets.

Revolution Big Data - Data Exploration Session 1

RevoScaleR options

Import and Explore Dow Jones Data

rxGetInfo Output

Extracting meta data about a variable using rxGetVarInfo()

Summarizing variables with rxSummary()

Revolution Big Data - Data Exploration Session 2

Exploring a Distribution with rxHistogram()

Plotting bivariate relationships with rxLinePlot()

Revolution Big Data - Data Exploration Session 3

Summarzing Variables with rxCrossTabs()

Summarzing Variables with rxCube()

Data Exploration

Chapter 2

Get introduced to the RevoScaleR functions used to manipulate and transform large datasets. Perform simple and complex transformations, and experiment with these via interactive exercises.

Revolution Big Data - Data Manipulation Session 1

Concatenating Multiple Datasets to a Single .xdf file

Converting From internal data.frame to external data.frame

Using rxDataStep() to Transform Data

Revolution Big Data - Data Manipulation Session 2

More Complex Transforms Using transformFuncs

Bad transformFunc functions

Data Manipulation

Chapter 3

Learn about the RevoScaleR functions used to analyze large datasets. Use these functions to perform analyses such as linear and logistic regression, k-means clustering, and decision tree estimation.

Data Preparation

Preparing Data For Analysis: Import

Preparing Data For Analysis: Exploration

Linear Model

Construct a linear model

Prediction

Generating Predictions and Residuals

Compute Standard Errors

Logistic Regression and the GLM

Logistic Regression

Individual Mortgage Information

Higher Default Probability

Individual Higher Default Probability

Valid Arguments for rxGLM

k Means

Computing k Means with rxKmeans()

Decision Trees

Create some Decision Trees

Data Analysis

Chapter 4

Revolution R Enterprise allows R users to process, visualize, and model terabyte-class data sets at a fraction of the time of legacy products without requiring expensive or specialized hardware. Introductory course for accomplished R users to experience the functionality of Revolution R Enterprise.<br><br> We like feedback; <a href="http://www.surveymonkey.com/s/AcademyRDataCamp_1" target = "_blank">Let us know how you experienced this course!</a>

Big Data Analysis with Revolution R Enterprise