Professionals are challenged each day by a changing landscape of technology and terminology. In recent history, especially the last 25 years there has been an explosion of terms and methods born that automate and improve decision-making and operations. One term called Analytics is an overarching description of a compilation of methodologies. But, AI (Artificial Intelligence), statistics, decision science, optimization which have been around for decades has resurged. Also, things like business intelligence, On-line Analytical Processing (OLAP) and many, many more have been born or reborn. How is someone to make sense of all this methodology, terminology? This book, the first in a series of three, provides a look at the foundations of artificial intelligence and analytics and why readers need an unbiased understanding of the subject. The authors include the basics such as algorithms, mental concepts, models, and paradigms in addition to the benefits of machine learning. The book also includes a chapter on data and the various forms of data. The authors wrap up this book with a look at next frontiers such as applications and designing your environment for success, which segue into the topics of the next two books in the series.
Author(s): Scott Burk; Gary D. Miner
Publisher: CRC Press
Year: 2020
Language: English
Pages: xxxvi+272
Cover
Half Title #2,0,-32767Title Page #4,0,-32767Copyright Page #5,0,-32767Table of Contents #6,0,-32767Foreword Number One #16,0,-32767Foreword Number Two #18,0,-32767Foreword Number Three #20,0,-32767Preface #22,0,-32767Endorsements
Authors #32,0,-32767Chapter 1 You Need This Book
Preamble
The Hip, the Hype, the Fears, the Intrigue, and the Reality:
Hype, Fear, and Intrigue No 1:
Hype, Fear, and Intrigue No 2:
Hype, Fear, and Intrigue No 3:
Professionals Need This Book
Introduction
Technology Keeps Raging, but We Need More Than Technology to Be Successful
Data and Analytics Explosion
A Bright Side of the Revolution
Where Is Someone to Turn for Information?
The Problem, Too Many Self-Interests: The Need for an Objective View
There Are Many Other Professional Stories That Are Concerned about Whether Analytics Is Important; Here Are a Few More Examples
What This Book Is Not:
Why This Book?
Sure, Business, but Why Healthcare, Public Policy, and Business?
How This Book Is Organized
References
Resources for the Avid Learner
Chapter 2 Building a Successful Program
Preamble
The Hip, the Hype, the Fears, the Intrigue, and the Reality
The Hype
Reality
The Hype
Reality
The Hype
Reality
Introduction
Culture and Organization – Gaps and Limitations
Gaps in Analytics Programs
Characterizing Common Problems
Don’t Confuse Organizational Gaps for Project Gaps
Justifying a Data-Driven Organization
Motivations
Critical Business Events
Analytics as a Winning Strategy
Part I – New Programs and Technologies
Part II – More Traditional Methods of Justification
Positive Return of Investment
Scale
Productivity
Reliability
Sustainability
Designing the Organization for Program Success
Motivation / Communication and Commitment
Establish Clear Business Outcomes
Organization Structure and Design
The Organization and Its Goals – Alignment
Organizational Structure
Centralized Analytics
Decentralized or Embedded Analytics
Multidisciplinary Roles for Analytics
Data Scientists
Data Engineers
Citizen Data Scientists
Developers
Business Experts
Business Leaders
Project Managers
Analytics Oversight Committee (AOC) and Governance Committee (Board Report)
Postscript
References
Resources for the Avid Learner
Chapter 3 Some Fundamentals – Process, Data, and Models
Preamble
The Hip, the Hype, the Fears, the Intrigue, and the Reality
The Hype
Reality
Introduction
Framework for Analytics – Some Fundamentals
Processes Drive Data
Models, Methods, and Algorithms
Models, Models, Models
Statistical Models
Rules of Thumb, Heuristic Models
A Note on Cognition
Algorithms, Algorithms, Algorithms
Distinction between Methods That Generate Models
There Is No Free Lunch
A Process Methodology for Analytics
CRISP-DM: The Six Phases:
Last Considerations
Data Architecture
Analytics Architecture
Postscript
References
Resources for the Avid Learner
Chapter 4 It’s All Analytics!
Preamble
Overview of Analytics – It’s All Analytics
Analytics of Every Form and Analytics Everywhere
Introduction
Analytics Mega List
Breaking it Down, Categorizing Analytics
Introduction
Gartner’s Classification
Descriptive Analytics
Diagnostic Analytics
Predictive Analytics
Prescriptive Analytics
Process Optimization
Some Additional Thoughts on Classifying Analytics
Fundamentals of Analytics – Data Basics
Introduction
Four Scales of Measurement
Data Formats
Data Stores
Provisioning Data for Analytics
Data Sourcing
Data Quality Assessment and Remediation
Integrate and Repeat
Exploratory Data Analysis (EDA)
Data Transformations
Data Reduction
Postscript
References
Resources for the Avid Learner
Chapter 5 What Are Business Intelligence (BI) and Visual BI?
Preamble
Introduction
Background and Chronology
Basic (Digital) Reporting
A View inside the Data Warehouse and Interactive BI
Beyond the Data Warehouse and Enhanced Interactive Visual BI and More
Business Activity Monitoring an Alert-Based BI, Version 4.0
Strengths and Weaknesses of BI
Transparency and Single Version of the Truth
Summary
Postscript
References
Resources for the Avid Learner
Chapter 6 What Are Machine Learning and Data Mining?
Preamble
Overview of Machine Learning and Data Mining
Is There a Difference?
A (Brief) Historical Perspective of Data Mining and Machine Learning
What Types of Analytics Are Covered by Machine Learning?
An Overview of Problem Types and Common Ground
The BIG Three!
Regression
Classification
Natural Language Processing (NLP)
Some (of Many) Additional Problem Classes
Association, Rules and Recommender Systems
Clustering
Some Comments on Model Types
Some Popular Machine Learning Algorithm Classes
Trees 1.0: Classification and Regression Trees or Partition Trees
Trees 2.0: Advanced Trees: Boosted Trees and Random Forests, for Classification and Regression
Regression Model Trees and Cubist Models
Logistic and Constrained/Penalized (LASSO, Ridge, Elastic Net) Regression
Multivariate Adaptive Regression Splines
Support Vector Machines (SVMs)
Neural Networks in 1000 Flavors
K-Means and Other Clustering Algorithms
Directed Acyclic Graph Analytics (Optimization, Social Networks)
Association Rules
AutoML (Automated Machine Learning)
Transparency and Processing Time of Algorithms
Model Use and Deployment
Major Components of the Machine Learning Process
Advantages and Limitations of Using Machine Learning
Postscript
References
Resources for the Avid Learner
Chapter 7 AI (Artificial Intelligence) and How It Differs from Machine Learning
Preamble
Introduction
Let Us Outline Two Types of AI Here – Weak AI and Strong AI
AI Background and Chronology
Short History of Digital AI
Resurrection in the 1980s
Beyond the Second AI Winter
Deep Learning, Bigger, and New Data
Next-Generation AI
Differences of BI, Data Mining, Machine Learning, Statistics vs AI
Strengths and Weakness
Some Weaknesses of AI
AI’s Future
“How ‘Rosy’ is the FUTURE for AI?”
Postscript
References
Resources for the Avid Learner
Chapter 8 What Is Data Science?
Preamble
Introduction
Mushing All the Terms – Same Thing?
Today’s Data Science?
Data Science vs BI and Data Scientist
Data Science vs Data Engineering vs Citizen Data Scientist
Backgrounds of Data Analytics Professionals
Young Professionals’ Input on What Makes a Great Data Scientist
Summary
Postscript
References
Resources for the Avid Learner
Chapter 9 Big Data and Bigger Data, Little Data, Cloud, and Other Data
Preamble
Introduction
Three Popular Forms and Two Divisions of Data
What Is Big Data?
Why the Push to Big Data? Why Is Big Data Technology Attractive?
The Hype of Big Data
Pivotal Changes in Big Data Technology
Brief Notes on Cloud
“Not Big Data” Is Alive and Well and Lessons from the Swamp
A Brief Note on Subjective and Synthetic Data
Other Important Data Focuses of Today and Tomorrow
Data Virtualization (DV)
Streaming Data
Events (Event-Driven or Event Data)
Geospatial
IoT (Internet of Things)
High-Performance In-Memory Computing Beyond Spark
Grid and GPU Computing
Near-Memory Computing
Data Fabric
Future Careers in Data
Postscript
References
For the Avid Learner
Chapter 10 Statistics, Causation, and Prescriptive Analytics
Preamble
Some Statistical Foundations
Introduction
Two Major Divisions of Statistics – Descriptive Statistics and Inferential Statistics
What Made Statistics Famous?
Criminal Trials and Hypothesis Testing
The Scientific Method
Two Major Paradigms of Statistics
Bayesian Statistics
Classical or Frequentist Statistics
Dividing It Up – Assumption Heavy and Assumption Light Statistics
Non-Parametric and Distribution Free Statistics (Assumption Light)
Four Domains in Statistics to Mention
Statistics in Predictive Analytics
Design of Experiments (DoE)
Statistical Process Control (SPC)
Time Series
An Ever-Important Reminder
Statistics Summary
Advantages of Statistics vs BI, Machine Learning and AI
Disadvantages of Statistics vs BI, Machine Learning and AI
Comparison of Data-Driven Paradigms Thus Far
Business Intelligence (BI)
Machine Learning and Data Mining
Artificial Intelligence (AI)
Statistics
Predictive Analytics vs Prescriptive Analytics – The Missing Link Is Causation
Assuming or Establishing Causation
Ladder of Causation
Predicting an Increasing Trend – Structural Causal Models and Causal Inference
Summary
Postscript
References
Resources for the Avid Learner
Chapter 11 Other Disciplines to Dive in Deeper: Computer Science, Management/Decision Science, Operations Research, Engineering (and More)
Preamble
Introduction
Computer Science
Management Science
Decision Science
Operations Research
Engineering
Finance and Econometrics
Simulation, Sensitivity and Scenario Analysis
Sensitivity Analysis
Scenario Analysis
Systems Thinking
Postscript
References
Resources for the Avid Learner
Chapter 12 Looking Ahead
Farewell, Until Next Time
