Heuristics in Analytics (SAS)

ISBN-13: 9781118347607 / Angielski / Twarda / 2014 / 256 str.

Carlos Andre Reis Pinheiro; Fiona McNeill

Opis

Employ heuristic adjustments for truly accurate analysis Heuristics in Analytics presents an approach to analysis that accounts for the randomness of business and the competitive marketplace, creating a model that more accurately reflects the scenario at hand. With an emphasis on the importance of proper analytical tools, the book describes the analytical process from exploratory analysis through model developments, to deployments and possible outcomes. Beginning with an introduction to heuristic concepts, readers will find heuristics applied to statistics and probability, mathematics, stochastic, and artificial intelligence models, ending with the knowledge applications that solve business problems. Case studies illustrate the everyday application and implication of the techniques presented, while the heuristic approach is integrated into analytical modeling, graph analysis, text analytics, and more. Robust analytics has become crucial in the corporate environment, and randomness plays an enormous role in business and the competitive marketplace. Failing to account for randomness can steer a model in an entirely wrong direction, negatively affecting the final outcome and potentially devastating the bottom line. Heuristics in Analytics describes how the heuristic characteristics of analysis can be overcome with problem design, math and statistics, helping readers to:

• Realize just how random the world is, and how unplanned events can affect analysis
• Integrate heuristic and analytical approaches to modeling and problem solving
• Discover how graph analysis is applied in real-world scenarios around the globe
• Apply analytical knowledge to customer behavior, insolvency prevention, fraud detection, and more
• Understand how text analytics can be applied to increase the business knowledge

Every single factor, no matter how large or how small, must be taken into account when modeling a scenario or event--even the unknowns. The presence or absence of even a single detail can dramatically alter eventual outcomes. From raw data to final report, Heuristics in Analytics contains the information analysts need to improve accuracy, and ultimately, predictive, and descriptive power.

A practical guide to deploying mathematical and statistical models when performing analytics The Heuristics in Analytics describes analytic processes and how they fit into the heuristic world around us.

Spis treści

Preface xi

Acknowledgments xix

About the Authors xxiii

Chapter 1: Introduction 1

The Monty Hall Problem 5

Evolving Analytics 8

Summary 18

Chapter 2: Unplanned Events, Heuristics, and the Randomness in Our World 23

Heuristics Concepts 26

The Butterfly Effect 30

Random Walks 37

Summary 44

Chapter 3: The Heuristic Approach and Why We Use It 45

Heuristics in Computing 47

Heuristic Problem–Solving Methods 51

Genetic Algorithms: A Formal Heuristic Approach 54

Summary 67

Chapter 4: The Analytical Approach 69

Introduction to Analytical Modeling 71

The Competitive–Intelligence Cycle 74

Summary 97

Chapter 5: Knowledge Applications That Solve Business Problems 101

Customer Behavior Segmentation 102

Collection Models 106

Insolvency Prevention 113

Fraud–Propensity Models 120

Summary 127

Chapter 6: The Graph Analysis Approach 129

Introduction to Graph Analysis 130

Summary 143

Chapter 7: Graph Analysis Case Studies 147

Case Study: Identifying Influencers in Telecommunications 149

Case Study: Claim Validity Detection in Motor Insurance 162

Case Study: Fraud Identification in Mobile Operations 178

Summary 188

Chapter 8: Text Analytics 191

Text Analytics in the Competitive–Intelligence Cycle 193

Linguistic Models 198

Text–Mining Models 200

Summary 207

Bibliography 209

Index 217

O autorze

CARLOS ANDRE REIS PINHEIRO is Visiting Professor at KU Leuven, Belgium. He headed the Analytical Lab at Oi in Brazil, one of the largest telecommunications companies in Latin America. Pinheiro has conducted Postdoctoral Research at Katholieke Universiteit Leuven, Belgium, Université de Savoie, France and Dublin City University, Ireland. He holds a PhD in Engineering from Federal University of Rio de Janeiro, Brazil. He worked at Brazil Telecom for almost ten years and also accomplished postdoctoral research at IMPA, Brazil, one of the most prestigious mathematical institutions in the world. He has published several papers in international journals and conferences and has four books (all in Portuguese) that focus on the internet, database, web warehousing, and analytical intelligence. He is the author of Social Network Analysis in Telecommunications, published by Wiley.

FIONA McNEILL has applied analytics to business problems since she began her career in 1992 and has consistently helped companies benefit from strategic use of data and analytics. Throughout her career, she has been affiliated with data and technology companies, from information and survey providers, IBM Global Services and for over fifteen years, at SAS. McNeill has published in academic journals, conducted education seminars and presented at both academic and industry conferences over the course of her career. She holds an M.A. in Quantitative Behavioral Geography from McMaster University, and graduated cum laude with a B.Sc. in Bio–Physical Systems, University of Toronto.

Blurb

Praise for Heuristics in Analytics: A Practical Perspective of What Influences Our Analytical World

Analytics is too often treated as a subject where the cleverness of computers generates the best answers. Fiona and Carlos return the focus to the cleverness of people, providing practical guidance about dealing with unknowns. Their text, with entertaining examples, puts the art back into the science of data mining.
Gordon S. Linoff, Founder, Data Miners, Inc., Author of Data Mining Techniques for Marketing, Sales, and Customer Relationship Management and Data Analysis Using SQL and Excel (Wiley)

With all the hype about big data, it s refreshing to find a book that discusses the practical aspects of analytics. Heuristics in Analytics makes a clear case for adding human experience and common sense to technology in order to solve real world business problems. Written in clear, non–mathematical language, the book explains how using heuristics together with analytics is often the fastest way to deliver decisions that are suitable for a specific use case, and quickly enough to fit the fast pace of business. The descriptions of heuristics concepts and guidance for how to use a heuristic approach to analytics should make this book a valuable addition to the manager s and practitioner s libraries.
Sue Feldman, Synthexis, Author of The Answer Machine

Heuristics in Analytics will provide patient readers with a second–to–none peek under the analytical tent. Analytics is no longer a black box as Dr. Pinheiro and Ms. McNeill describe with surgical precision processes associated with defining problems, describing scenarios, developing and deploying models and selecting appropriate tools and techniques. The case studies are powerful and illuminating. There is much truth talking in these pages.
Thornton May, Futurist & Author of The New Know: Innovation Powered by Analytics