Machine Learning for Auditors

Use artificial intelligence (AI) techniques to build tools for auditing your organization. This is a practical book with implementation recipes that demystify AI, ML, and data science and their roles as applied to auditing. You will learn about data analysis techniques that will help you gain insights into your data and become a better data storyteller. The guidance in this book around applying artificial intelligence in support of audit investigations helps you gain credibility and trust with your internal and external clients. A systematic process to verify your findings is also discussed to ensure the accuracy of your findings.

Machine Learning for Auditors provides an emphasis on domain knowledge over complex data science know how that enables you to think like a data scientist. The book helps you achieve the objectives of safeguarding the confidentiality, integrity, and availability of your organizational assets. Data science does not need to be an intimidating concept for audit managers and directors. With the knowledge in this book, you can leverage simple concepts that are beyond mere buzz words to practice innovation in your team. You can build your credibility and trust with your internal and external clients by understanding the data that drives your organization.

What You Will Learn

• Understand the role of auditors as trusted advisors
• Perform exploratory data analysis to gain a deeper understanding of your organization
• Build machine learning predictive models that detect fraudulent vendor payments and expenses
• Integrate data analytics with existing and new technologies
• Leverage storytelling to communicate and validate your findings effectively
• Apply practical implementation use cases within your organization

Who This Book Is For

AI Auditing is for internal auditors who are looking to use data analytics and data science to better understand their organizational data. It is for auditors interested in implementing predictive and prescriptive analytics in support of better decision making and risk-based testing of your organizational processes. 

Table of Contents

Part I: Trusted Advisors

Chapter 1: Three Lines of Defense

Chapter 2: Common Audit Challenges

Chapter 3: Existing Solutions

Chapter 4: Data Analytics

Chapter 5: Analytics Structure and Environment

Part II: Understanding Artificial Intelligence

Chapter 6: Introduction to Al, Data Science, and Machine Learning

Chapter 7: Myths and Misconceptions

Chapter 8: Trust, but Verify

Chapter 9: Machine Learning Fundamentals

Chapter 10: Data Lakes

Chapter 11: Leveraging the Cloud

Chapter 12: SCADA and Operational Technology

Part III: Storytelling

Chapter 13: What Is Storytelling?

Chapter 14: Why Storytelling?

Chapter 14: Why Storytelling?

Chapter 15: When to Use Storytelling?

Chapter 16: Types of Visualizations

Chapter 17: Effective Stories

Chapter 18: Storytelling Tools

Chapter 19: Storytelling in Auditing

Part IV: Implementation Recipes

Chapter 20: How to Use the Recipes

Chapter 21: Fraud and Anomaly Detection

Chapter 22: Access Management

Chapter 23: Project Management

Chapter 24: Data Exploration

Chapter 25: Vendor Duplicate Payments

Chapter 26: CMTs 2.0

Chapter 27: Log Analysis

Chapter 28: Concluding Remarks

About the Author

​Maris Sekar is a professional computer engineer, Certified Information Systems Auditor (ISACA), and Senior Data Scientist (Data Science Council of America). He has a passion for using storytelling to communicate on high-risk items within an organization to enable better decision making and drive operational efficiencies. He has cross-functional work experience in various domains such as risk management, data analysis and strategy, and has functioned as a subject matter expert in organizations such as PricewaterhouseCoopers LLP, Shell Canada Ltd., and TC Energy. Maris’ love for data has motivated him to win awards, write LinkedIn articles, and publish two papers with IEEE on applied machine learning and data science.