Data Mining and Predictive Analytics –
21st Century’s Most Powerful New Management Tools

      

If you’re preparing to lead or participate in a data analytics initiative,
this is the one book you must read!



Jeff Deal

Jeff Deal is the Vice President of Operations for Elder Research with more than thirty years experience managing healthcare and consulting businesses. Though Jeff has been deeply immersed in the analytics business in a leadership roles, he well remembers his initial introduction to this rapidly-moving field, and he brings that layman’s perspective to the book. Jeff enjoys speaking on the subject of organizational challenges to meeting data analytics goals, and he chairs the annual Predictive Analytics World – Healthcare conference, which attracts leading analytics professionals in the healthcare industry from around the country.

He holds a Master of Health Administration degree from Virginia Commonwealth University in Richmond, Virginia, and a Bachelor of Arts degree from the College of William and Mary in Williamsburg, Virginia, where he was a member of the wrestling team. Jeff and his wife Jennifer have four children. In his spare time he enjoys hiking, reading, and an increasing amount of travel now that his kids have all moved out of the house.


Gerhard Pilcher

Gerhard Pilcher is Chief Executive Officer for Elder Research with more than thirty years of analytics experience in the commercial businesses and government institutions in the United States and abroad. He is an adjunct faculty member in the Math and Statistics Masters program at Georgetown University and a regular instructor in the SAS Business Knowledge Series courses Gerhard currently serves on the advisory boards of the Institute for Advanced Analytics and George Washington University Master of Science in Business Analytics program.

Gerhard earned a Master of Science degree in analytics from the Institute for Advanced Analytics at North Carolina State University in Raleigh, North Carolina. Gerhard and his wife, Denise, have two children. In his spare time, he especially enjoys outdoor activities, including mountaineering and trail running.


A Must Read for Every Executive!

In this practical guide for organizational leaders and top-level executives, industry experts Jeff Deal and Gerhard Pilcher explain in clear, understandable English…

 

  • What data mining and predictive analytics are
  • Why they are such powerful management tools
  • How to establish and manage a data science service

 

Complete with solid advice and instructive case studies, this book demonstrates how to harness the power of data mining and predictive analytics, and avoid costly mistakes.

Use it to gain a quick overview of analytics and as a handy resource to be referred to during a project.

If you're preparing to lead or participate in a data analytics initiative, this is the one book you must read!


CONTENTS

Foreword

Introduction

Introduction and Overview

  1. Empowering the Decision Makers
    • Hunting for Needles in Haystacks
    • Breaking the Mind Barrier
    • A Variety of Applications
  2. Clearing Up the Confusion
    • Four Categories of Modeling Technology
    • Supervised vs. Unsupervised Learning
    • Ten Levels of Analytics
    • Levels and Advanced Data Types

The Analytic Organization

  1. Leading a Data Analytics Initiative
    • Starting Small
    • Cultivating the Culture
    • Managing a Data Analytics Initiative
    • An Illustrative Example
    • Leadership is Key
    • More Illustrative Examples
  1. Staffing a Data Analytics Project
    • Individual or Team?
    • Assembling the Team
    • What is a Data Scientist?
    • More than Academic Credentials
    • The Most Important Quality
    • Mike Thurber’s Story
    • Building Teams through “Gap Analysis”
  1. Acquiring the Right Tools
    • A Variety of Techniques and Disciplines
    • Levels of Tools
    • Sources of Tools
    • A Word about Open-Source Tools
    • Trends
  1. Hiring Data Analytics Consultants
    • Discerning Fact from Hype
    • Evaluating Industry Experience
    • Evaluating Analytics Experience
    • Finding the Right Consultant

The Modeling Process

  1. Understanding the Data Mining Process
    • The CRISP-DM Process
    • Resisting the Temptation to Take Shortcuts
  1. Understanding the Business
    • Clarifying Your Objective
    • Defining the Terminology
    • Framing the Questions
  1. Understanding and Preparing the Data
    • Preparing Data
    • Cleaning Data
    • “Perfect” Data
    • Collecting and Preparing the Data
    • Fostering Cooperation
    • Data Governance
  1. Building the Model
    • Inside the “Black Box”
    • An Illustrative Example
    • Choosing a Model
    • Response Surfaces of Predictive Models
    • The Tradeoff between Accuracy and Interpretability
    • Dealing with Variance
    • Model Ensembles
  1. Validating the Model
    • Technical Validation
    • Checking for Mistakes
    • Checking for Generalization
    • Using Experts to Qualify Model Results
    • Target Shuffling
    • Business Validation
  1. Deploying the Model
    • Planning and Budgeting for Deployment
    • Business Processes Are Key
    • Illustrative Examples
    • Four Important Questions
  1. Realizing the Transformation
    • Realizing the Potential
    • The Tipping Point

 

Appendix

  • Are Orange Cars Really Least Likely To Be Lemons?
  • Additional Resources

About the Authors

About Elder Research, Inc.

Index

Authors: Jeff Deal & Gerhard Pilcher
Publisher: Data Science Publishing
Language: English
ISBN: 978-0-9967121-0-1
Pages: 184
Trim Size: 6 x 9
Genre: Business
Available as:

  • Soft cover
  • e-book
Price (soft cover:) $19.95

Available nationally at bookstores, libraries
and your favorite online bookstore


Reviews

Delivers the two ingredients you need for success: 1) an understanding of the technology so you can speak the quants’ language and 2) a guide to analytics management best practices, including how to build your analytics team and avert the most costly pitfalls.” (Excerpt from the Foreword)

Eric Siegel, Ph.D., founder of Predictive Analytics World, Author of Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die

Captures all the critical elements and decades of experiences into a few clear pages that will light the path for predictive improvements. I will be sharing it with my leadership and program managers. Great job gentlemen in making a complex equation simple to follow!

Fred Walker, Technical Director Counterintelligence, National Security Agency

Amidst the concerns about the shortage of data scientists, a larger, over¬looked obstacle is finding C-, VP-, and director-level leaders who understand enough about advanced analytics to hire, manage, and deploy solutions. Deal and Pilcher have written a practical and insightful ‘primer for executives’ to expertly fill this void.

Dean Abbott, Co-Founder and Chief Data Scientist at SmarterHQ, Author of Applied Predictive Analytics

Data science and big data often don’t live up to their silver bullet hype. Why? Because IT and business are so different, and so hard to harmonize. This book is an excellent remedy; Deal and Pilcher distill a decade of analytics experience into a vital guide to what works. Mining Your Own Business is a must read for anyone interested in being right—by harnessing data to drive decisions.

Peter Aiken, PhD Founding Director/Data Blueprint Associate Professor of Information Systems/Virginia Commonwealth University

Deal and Pilcher have distilled their decades of experience into an easy-to-read book that will benefit any business person dealing with analytics. They keep technical details to a minimum while focusing on the key facts, decisions, and actions that business people need to be successful with analytics. Abundant real-world examples reinforce their practical and valuable advice. Your time reading the book will be well spent!

Bill Franks, Chief Analytics Officer, Teradata, Author of Taming The Big Data Tidal Wave and The Analytics Revolution

Jeff and Gerhard have taken their in-depth academic and seasoned business knowledge of analytics, and distilled it into an engaging and enlightening book that is both a “how-to” guide and a reference source. Using materials from a varied client base ranging from government agencies to hedge funds, they clearly outline the steps and best practices to develop leading-edge analytics programs. They describe sophisticated techniques and provide illustrative examples of the resulting quantifiable value. Whether you are CIO or a non-technical senior executive, this book will help you answer two key questions: How can data mining and predictive analytics advance my business? And perhaps even more importantly, 'How do I get started?'

Rafael Pabón Consultant and CIO, Sherman Hill Group

I think this book could be a game-changer! Every business and government leader who has anything to do with data analytics should read it.

Yoony Doh, Advanced Analytics Practice Area Lead and Project Manager Halfaker and Associates, LLC

Resources

(Click on the title to read and download the PDF.)

The Ten Levels of Analytics: Every technical project involves some sort of analytics, ranging from simply reporting key facts, to predicting new events. This eBook defines ten increasingly sophisticated levels of analytics so that teams can assess where they stand and to what they aspire.

Along the way, definitions of three types of analytic inquiry and four categories of modeling technology are clarified. The eBook illustrates these levels with examples using tabular data representations commonly found in spreadsheets and single database tables.

As the ability to collect and fuse data from different sources increases, advanced data types such as time series, spatial data, and graph data are moving into the analytic mainstream. In the second portion of this eBook, the Levels are extended to encompass these emerging data types, providing data complexity as second dimension for categorization alongside algorithmic sophistication.

The Ten Most Common Data Mining Business Mistakes: The results of data mining applications can have big payoffs, but the implementation of data mining techniques can also present substantial challenges. Many companies fail to reap the benefits because they make crucial mistakes in planning and deployment. This paper discusses the ten business mistakes that frequently cause data science projects to fall short of expectations. Awareness of these common mistakes will better equip organizational leaders to plan and guide data mining engagements to successful conclusions.

Operationalizing Analytics Solutions and Methods: It is a challenge to make an analytic model work in a production environment, as it requires teamwork from IT, Data Management, Analytics, and Business units. The goals of the process are threefold:

  1. Build models with repeatable, reliable results that do not depend on any single person or working environment to operate.
  2. Make model results available to end-users in a timely and useable manner.
  3. Monitor model performance on an ongoing basis to ensure quality and alert analysts to any degradation over time.

This white paper reviews strategies for achieving these goals.