Introduction to Prescriptive AI: A Primer for Decision Intelligence Solutioning with Python

Gain a working knowledge of prescriptive AI, its history, and its current and future trends. This book will help you evaluate different AI-driven predictive analytics techniques and help you incorporate decision intelligence into your business workflow through real-world examples. The book kicks off with an introduction to decision intelligence and provides insight into prescriptive AI and how it can be woven into various business strategies and frameworks. You'll then be introduced to different decision intelligence methodologies and how to implement them, along with advantages and limitations of each. Digging deeper, the authors then walk you through how to perform simulations and interpret the results. A full chapter is devoted to embedding decision intelligence processes and outcomes into your business workflow using various applications. The book concludes by exploring different cognitive biases humans are prone to, and how those biases can be eliminated by combining machine and human intelligence. Upon completing this book, you will understand prescriptive AI, tools, and techniques and will be ready to incorporate them into your business workflow. What You Will Learn Implement full-fledged decision intelligence applications using Python Leverage the tools, techniques, and methodologies for prescriptive AI Understand how prescriptive AI can be used in different domains through practical examples Interpret results and integrate them into your decision making Who This Book Is For Data Scientists and Machine Learning Engineers, as well as business professionals who want to understand how AI-driven decision intelligence can help grow their business. Table of Contents 5 About the Authors 9 About the Technical Reviewer 11 Acknowledgments 12 Introduction 13 Chapter 1: Decision Intelligence Overview 15 Types of AI 16 Decision Intelligence 19 Decision Intelligence History 20 Challenges in AI Adoption 21 How Can DI Help Bridge the Gap Between AI and Business? 22 The Need for Decision Intelligence 23 The Evolution of Decision-Making 25 Challenges 27 Applications 28 Understanding Where Decision Intelligence Fits Within the AI Life Cycle 30 Decision Intelligence Methodologies 32 Some Potential Pros and Cons of DI 36 Examples of How Companies Are Leveraging DI 38 Conclusion 39 Chapter 2: Decision Intelligence Requirements 40 Why Do AI Projects Fail? 40 DI Requirements Framework 43 Planning 44 Approach 46 Approval Mechanism/Organization Alignment 49 Key Performance Indicators 50 Define Clear Metrics 51 Value 53 Return on Investment 54 Value per Decision 55 Consumption of the AI Predictions 56 Conclusion 59 Chapter 3: Decision Intelligence Methodologies 60 Decision-Making 60 Types of Decision-Making 61 Individual vs. Group Decision-Making 61 Single- vs. Multiple-Criterion Decision-Making 62 Strategic, Tactical, and Operational Decision-Making 63 Decision-Making Process 64 Decision-Making Process Example 67 Decision-Making Methodologies 70 Human-Only Decision-Making 71 Random Decisions 72 Morality/Ethics Based 73 Experience Based 75 Authority Based 76 Consensus Based 78 Voting Based 79 Threshold Based 81 First Acceptable Match Based 82 Optimization/Maximization Based 84 Cognitive Bias Due to Human-Only Decision-Making 86 Human-Machine Decision-Making 86 Instruction/Rule-Based Systems 87 Mathematical Models 88 Probabilistic Models 90 AI-Based Models 92 Machine-Only Decision-Making 94 Autonomous Systems 94 Conclusion 96 Chapter 4: Interpreting Results from Different Methodologies 97 Decision Intelligence Methodology: Mathematical Models 97 Linear Models 98 Nonlinear Models 102 Decision Intelligence Methodology: Probabilistic Models 106 Markov Chain 107 Decision Intelligence Methodology: AI/ML Models 115 Conclusion 120 Chapter 5: Augmenting Decision Intelligence Results into the Business Workflow 121 Challenges 122 Workflow 124 Decision Intelligence Apps 126 How and Why? 127 User-Friendly Interfaces 128 Augmenting AI Predictions to Business Workflow 130 Connect to Business Tools 132 Map the Data 133 Conclusion 134 Chapter 6: Actions, Biases, and Human-in-the-Loop 136 Key Ethical Considerations in AI 136 Actions, Biases, and Human-in-the-Loop 138 Cognitive Biases 139 Why Is Detecting Bias Important? 140 Types 141 What Happens If Bias Is Ignored? 143 Bias Detection 144 What Do Bias Tools Do? 145 Incorporation of Feedback Through Human Intervention 147 How to Build HITL Systems? 150 Example: Customer Churn 151 Conclusion 152 Chapter 7: Case Studies 154 Case Study 1: Telecom Customer Churn Management 154 Case Study 2: Mobile Phone Pricing/Configuration Strategy 184 Conclusion 199 Index 200