Book Image

Mastering Predictive Analytics with R - Second Edition

By : James D. Miller, Rui Miguel Forte
Book Image

Mastering Predictive Analytics with R - Second Edition

By: James D. Miller, Rui Miguel Forte

Overview of this book

R offers a free and open source environment that is perfect for both learning and deploying predictive modeling solutions. With its constantly growing community and plethora of packages, R offers the functionality to deal with a truly vast array of problems. The book begins with a dedicated chapter on the language of models and the predictive modeling process. You will understand the learning curve and the process of tidying data. Each subsequent chapter tackles a particular type of model, such as neural networks, and focuses on the three important questions of how the model works, how to use R to train it, and how to measure and assess its performance using real-world datasets. How do you train models that can handle really large datasets? This book will also show you just that. Finally, you will tackle the really important topic of deep learning by implementing applications on word embedding and recurrent neural networks. By the end of this book, you will have explored and tested the most popular modeling techniques in use on real- world datasets and mastered a diverse range of techniques in predictive analytics using R.

Related Content you might be interested in

Current Title:

Small book image
Mastering Predictive Analytics with R - Second Edition
James D. Miller | Rui Miguel Forte
Aug, 2017 | 448 pages
Small book image
Statistics for Data Science
James D Miller
Nov, 2017 | 286 pages
Small book image
Mastering Machine Learning with R
Cory Lesmeister
Apr, 2017 | 420 pages
Small book image
Mastering Machine Learning with R.
Cory Lesmeister
Jan, 2019 | 354 pages
Table of Contents (22 chapters)
Mastering Predictive Analytics with R Second Edition
Mastering Predictive Analytics with R Second Edition
Credits
Credits
About the Authors
About the Authors
About the Reviewer
About the Reviewer
www.PacktPub.com
www.PacktPub.com
Customer Feedback
Customer Feedback
Preface
Preface
Free Chapter
1
Gearing Up for Predictive Modeling
Gearing Up for Predictive Modeling
Models
Types of model
The process of predictive modeling
Summary
2
Tidying Data and Measuring Performance
Tidying Data and Measuring Performance
Getting started
Tidying data
Categorizing data quality
Performance metrics
Cross-validation
Learning curves
Summary
3
Linear Regression
Linear Regression
Introduction to linear regression
Simple linear regression
Multiple linear regression
Assessing linear regression models
Problems with linear regression
Feature selection
Regularization
Polynomial regression
Summary
4
Generalized Linear Models
Generalized Linear Models
Classifying with linear regression
Introduction to logistic regression
Predicting heart disease
Assessing logistic regression models
Regularization with the lasso
Classification metrics
Extensions of the binary logistic classifier
Poisson regression
Negative Binomial regression
Summary
5
Neural Networks
Neural Networks
The biological neuron
The artificial neuron
Stochastic gradient descent
Multilayer perceptron networks
The back propagation algorithm
Predicting the energy efficiency of buildings
Predicting glass type revisited
Predicting handwritten digits
Radial basis function networks
Summary
6
Support Vector Machines
Support Vector Machines
Maximal margin classification
Support vector classification
Kernels and support vector machines
Predicting chemical biodegration
Predicting credit scores
Multiclass classification with support vector machines
Summary
7
Tree-Based Methods
Tree-Based Methods
The intuition for tree models
Algorithms for training decision trees
Predicting class membership on synthetic 2D data
Predicting the authenticity of banknotes
Predicting complex skill learning
Improvements to the M5 model
Summary
8
Dimensionality Reduction
Dimensionality Reduction
Defining DR
Summary
9
Ensemble Methods
Ensemble Methods
Bagging
Boosting
Predicting atmospheric gamma ray radiation
Predicting complex skill learning with boosting
Summary
10
Probabilistic Graphical Models
Probabilistic Graphical Models
A little graph theory
Bayes' theorem
Conditional independence
Bayesian networks
The Naïve Bayes classifier
Summary
11
Topic Modeling
Topic Modeling
An overview of topic modeling
Latent Dirichlet Allocation
Modeling the topics of online news stories
Modeling tweet topics
Summary
12
Recommendation Systems
Recommendation Systems
Rating matrix
Collaborative filtering
Singular value decomposition
Predicting recommendations for movies and jokes
Loading and pre-processing the data
Exploring the data
Other approaches to recommendation systems
Summary
13
Scaling Up
Scaling Up
Starting the project
Characteristics of big data
Training models at scale
A path forward
Alternatives
Summary
14
Deep Learning
Deep Learning
Machine learning or deep learning
What is deep learning?
Summary
Index
Index

About the Authors

James D. Miller is an IBM-certified expert, creative innovator, accomplished director, senior project leader, and application/system architect. He has over 35 years of extensive experience in application and system design and development across multiple platforms and technologies. His experience includes introducing customers to new technologies and platforms, integrating with IBM Watson Analytics, Cognos BI, and TM1. He has worked in web architecture design, systems analysis, GUI design and testing, database modeling, systems analysis, design and development of OLAP, web and mainframe applications and systems utilization, IBM Watson Analytics, IBM Cognos BI and TM1 (TM1 rules, TI, TM1Web, and Planning Manager), Cognos Framework Manager, dynaSight - ArcPlan, ASP, DHTML, XML, IIS, MS Visual Basic and VBA, Visual Studio, Perl, Splunk, WebSuite, MS SQL Server, Oracle, and Sybase server. James's responsibilities have also included all aspects of Windows and SQL solution development and design, such as analysis; GUI (and website) design; data modeling; table, screen/form, and script development; SQL (and remote stored procedures and triggers) development/testing; test preparation; and management and training of programming staff.

His other experience includes the development of ETL infrastructures, such as data transfer automation between mainframe (DB2, Lawson, Great Plains, and so on) system and client/server SQL Server, web-based applications, and the integration of enterprise applications and data sources. James has been a web application development manager responsible for the design, development, QA, and delivery of multiple websites, including online trading applications and warehouse process control and scheduling systems, as well as administrative and control applications. He was also responsible for the design, development, and administration of a web-based financial reporting system for a 450-million dollar organization, reporting directly to the CFO and his executive team.

Furthermore, he has been responsible for managing and directing multiple resources in various management roles, including as project and team leader, lead developer, and application development director. James has authored Cognos TM1 Developers Certification Guide, Mastering Splunk, and a number of white papers on best practices, including Establishing a Center of Excellence. He continues to post blogs on a number of relevant topics based on personal experiences and industry best practices. James is a perpetual learner, continuing to pursue new experiences and certifications. He currently holds the following technical certifications: IBM Certified Business Analyst - Cognos TM1 IBM Cognos TM1 Master 385 Certification (perfect score of 100%), IBM Certified Advanced Solution Expert - Cognos TM1, IBM Cognos TM1 10.1 Administrator Certification C2020-703 (perfect score of 100%), IBM OpenPages Developer Fundamentals C2020-001-ENU (98% in exam), IBM Cognos 10 BI Administrator C2020-622 (98% in exam), and IBM Cognos 10 BI Professional C2020-180.

He specializes in the evaluation and introduction of innovative and disruptive technologies, cloud migration, IBM Watson Analytics, Cognos BI and TM1 application design and development, OLAP, Visual Basic, SQL Server, forecasting and planning, international application development, business intelligence, project development and delivery, and process improvement.

I'd like to thank, Nanette L. Miller, and remind her that "Your destiny is my destiny. Your happiness is my happiness." I'd also like to thank Shelby Elizabeth and Paige Christina, who are both women of strength and beauty and whom I have no doubt will have a lasting, loving effect on the world.

Rui Miguel Forte is currently the chief data scientist at Workable. He was born and raised in Greece and studied in the UK. He is an experienced data scientist, with over 10 years of work experience in a diverse array of industries spanning mobile marketing, health informatics, education technology, and human resources technology. His projects have included predictive modeling of user behavior in mobile marketing promotions, speaker intent identification in an intelligent tutor, information extraction techniques for job applicant resumes, and fraud detection for job scams. He currently teaches R, MongoDB, and other data science technologies to graduate students in the Business Analytics MSc program at the Athens University of Economics and Business. In addition, he has lectured at a number of seminars, specialization programs, and R schools for working data science professionals in Athens.

His core programming knowledge is in R and Java, and he has extensive experience of a variety of database technologies, such as Oracle, PostgreSQL, MongoDB, and HBase. He holds a master's degree in Electrical and Electronic Engineering from Imperial College London and is currently researching machine learning applications in information extraction and natural language processing.

Behind every great adventure is a good story and writing a book is no exception. Many people contributed to making this book a reality. I would like to thank the many students I have taught at AUEB whose dedication and support has been nothing short of overwhelming. They should be rest assured that I have learned just as much from them as they have learned from me, if not more. I also want to thank Damianos Chatziantoniou for conceiving a pioneering graduate data science program in Greece. Workable has been a crucible for working alongside incredibly talented and passionate engineers on exciting data science projects that help businesses around the globe. For this, I would like to thank my colleagues and in particular the founders, Nick and Spyros, who created a diamond in the rough. I would like to thank Subho, Govindan, and all the folks at Packt for their professionalism and patience. My family and extended family have been an incredible source of support on this project. In particular, I would like to thank my father, Libanio, for inspiring me to pursue a career in the sciences and my mother, Marianthi, for always believing in me far more than anyone else ever could. My wife, Despoina, patiently and fiercely stood by my side even as this book kept me away from her during her first pregnancy. Last but not least, my baby daughter slept quietly and kept a cherubic vigil over her father during the book review phase. She helped me in ways words cannot describe.

Previous Section
Next Chapter