0%

Explore the most serious prevalent ethical issues in data science with this insightful new resource

The increasing popularity of data science has resulted in numerous well-publicized cases of bias, injustice, and discrimination. The widespread deployment of “Black box” algorithms that are difficult or impossible to understand and explain, even for their developers, is a primary source of these unanticipated harms, making modern techniques and methods for manipulating large data sets seem sinister, even dangerous. When put in the hands of authoritarian governments, these algorithms have enabled suppression of political dissent and persecution of minorities. To prevent these harms, data scientists everywhere must come to understand how the algorithms that they build and deploy may harm certain groups or be unfair.

Responsible Data Science delivers a comprehensive, practical treatment of how to implement data science solutions in an even-handed and ethical manner that minimizes the risk of undue harm to vulnerable members of society. Both data science practitioners and managers of analytics teams will learn how to:

  • Improve model transparency, even for black box models
  • Diagnose bias and unfairness within models using multiple metrics
  • Audit projects to ensure fairness and minimize the possibility of unintended harm

Perfect for data science practitioners, Responsible Data Science will also earn a spot on the bookshelves of technically inclined managers, software developers, and statisticians.

Table of Contents

  1. Cover
  2. Title Page
  3. Introduction
    1. What This Book Covers
    2. Who Will Benefit Most from This Book
    3. Looking Ahead in This Book
    4. Special Features
    5. Code Repository
  4. Part I: Motivation for Responsible Data Science and Background Knowledge
    1. CHAPTER 1: Responsible Data Science
    2. The Optum Disaster
    3. Jekyll and Hyde
    4. Eugenics
    5. Ethical Problems in Data Science Today
    6. Predictive Models
    7. Two Opposing Forces
    8. Summary
    9. CHAPTER 2: Background: Modeling and the Black-Box Algorithm
    10. Assessing Model Performance
    11. Intrinsically Interpretable Models vs. Black-Box Models
    12. Summary
    13. CHAPTER 3: The Ways AI Goes Wrong, and the Legal Implications
    14. AI and Intentional Consequences by Design
    15. The Legal and Regulatory Landscape around AI
    16. Summary
    17. Notes
  5. Part II: The Responsible Data Science Process
    1. CHAPTER 4: The Responsible Data Science Framework
    2. Why We Keep Building Harmful AI
    3. The Face Thieves
    4. An Anatomy of Modeling Harms
    5. Efforts Toward a More Responsible Data Science
    6. Summary
    7. Notes
    8. CHAPTER 5: Model Interpretability: The What and the Why
    9. The Sexist Résumé Screener
    10. The Necessity of Model Interpretability
    11. Connections Between Predictive Performance and Interpretability
    12. Uniting (High) Model Performance and Model Interpretability
    13. Real-World Successes of Interpretability Methods
    14. Addressing Critiques of Interpretability Methods
    15. The Forking Paths of Model Interpretability
    16. The Four-Measure Baseline
    17. Building Our Own Credit Scoring Model
    18. Summary
    19. Notes
  6. Part III: RDS in Practice
    1. CHAPTER 6: Beginning a Responsible Data Science Project
    2. How the Responsible Data Science Framework Addresses the Common Cause
    3. Datasets Used
    4. Common Elements Across Our Analyses3
    5. Beginning a Responsible Data Science Project
    6. Summary
    7. Notes
    8. CHAPTER 7: Auditing a Responsible Data Science Project
    9. Fairness and Data Science in Practice1
    10. Classification Example: COMPAS
    11. Summary
    12. Notes
    13. CHAPTER 8: Auditing for Neural Networks
    14. Why Neural Networks Merit Their Own Chapter1
    15. Beginning a Responsible Neural Network Project
    16. Auditing Neural Networks for Natural Language Processing
    17. Summary
    18. Notes
    19. CHAPTER 9: Conclusion
    20. How Can We Do Better?
    21. A Better Future If We Can Keep It
    22. Note
  7. Index
  8. Copyright
  9. About the Authors
  10. About the Technical Editor
  11. Acknowledgments
  12. End User License Agreement
44.200.169.91