Machine Learning Advances in Payment Card Fraud Detection

Erscheinungsjahr 2019
ISBN: 978-0-12-813415-3
Verlag: Elsevier Science

Buch, Englisch, 350 Seiten, Format (B × H): 152 mm x 229 mm, Gewicht: 550 g

Machine Learning Advances in Payment Card Fraud Detection provides a thorough review of the state-of-the-art in fraud detection research that is ideal for graduate-level readers and professionals. Through a comprehensive examination of fraud analytics that covers data collection, steps for cleaning and processing data, tools for analysing data, and ways to draw insights, the book argues for a new direction to be taken in developing state-of the-art payment fraud detection techniques. It uses an extensive analysis and description of an exemplar fraud detection algorithm, SOAR, to illustrate how a detailed understanding of the payment fraud domain can be used to motivate further advances in fraud detection techniques. The book concludes with a discussion of opportunities for future research, such as developing holistic approaches for countering fraud.



- Provides a detailed analysis of the payment fraud detection domain
- Develops an evaluation methodology for fraud detection techniques that takes full account of the business needs of the payment card industry
- Introduces state-of the-art payment fraud detection techniques

Zielgruppe

<p>Graduate level (MBA) and professionals working in credit card fraud detection and analysis</p>

Autoren/Hrsg.

Ryman-Tubb, Nick

Krause, Paul

Fachgebiete

• Wirtschaftswissenschaften Finanzsektor & Finanzdienstleistungen Unternehmensfinanzierung
• Wirtschaftswissenschaften Betriebswirtschaft Unternehmensfinanzen

Inhaltsverzeichnis

1. Introduction
2. History of payment cards
3. Growth of payment card fraud
4. The fraud detection problem
5. Understanding the fraud domain
6. Cost of payment fraud
7. History of payment card fraud detection methods
8. The pivotal event and disruptive technologies
9. The Sparse Oracle-based Adaptive Rule extraction algorithm
10. Real-world data empirical evaluation
11. Discussion and conclusion

Über Autor(innen)

Ryman-Tubb, Nick
Nick F. Ryman-Tubb helped to pioneer the application of artificial intelligence (AI) and deep learning neural networks within the financial industry. In 1986, he founded Neural Technologies in the UK, among the first AI businesses focused on risk, banking, and payment fraud. After his exit in 2000, Nick joined businesses that today deploy his AI in insurance, money laundering, contactless/mobile payment fraud detection, protecting over 150 institutions, more than 3 million merchants, 1 billion cards, and over 30 billion credit/debit card transactions a year. Nick is a professor and Machine Learning Impresario at the University of Surrey where he teaches and continues his research. He recently formed the Institute for Financial Innovation in Transactions and Security (FITS) as a non-profit organisation with a simple vision - to dramatically reduce payment fraud using AI. Today, FITS works with all its industry members towards this shared vision.