Credit Card Fraud – Why the Database Matters
Learn how databases combat credit card fraud at Big Data LDN 2019. Expert insights from Dheeraj Mudgil on data security and fraud prevention.
Course Content
Credit Card Fraud – Why the Database Matters is an intermediate-level Pluralsight course taught by Big Data LDN. It runs 25m, is taught in English, and includes a certificate of completion. It is a paid course, first published June 2021.
Fraud Detection at Scale: How Barclays Optimized its System
Barclays, a major financial institution processing over 30 million payment transactions daily for 20 million customers, faced significant challenges in maintaining effective transaction fraud detection. Their existing system relied on numerous, complex, and bespoke engineering solutions across various business units. This fragmented approach hindered scalability and posed a major obstacle to strategic goals.
The core problem stemmed from the inability to seamlessly re-use large-scale user profile datasets. This resulted in lengthy end-to-end fraud detection processes, introducing risks such as:
- Stand-in Processing (STIP): Increased processing time creates vulnerabilities.
- Data Consistency Issues: Inaccurate or conflicting data leads to errors.
- False Positives & False Negatives: Impaired accuracy in identifying fraudulent transactions.
Expert analysis pinpointed a limiting database deployment and technology as the root cause of these issues. To overcome these hurdles, Barclays implemented a new database technology, resulting in a significantly improved fraud detection system.
The new solution delivered substantial improvements:
- Scalability: The Barclays dataset was scaled from 3TB to 30TB+ within three years.
- Rule Sharing: Fraud rules could be consistently shared across all platforms.
- Machine Learning Facilitation: Enhanced capabilities for consistent machine learning applications.
- Ultra-Low Latency: Achieved a maximum of two-digit (<100) millisecond response time for the 99.99th percentile of transactions – crucial for real-time fraud prevention.
This case study highlights the critical role of a robust database in achieving efficient and effective payment fraud detection, especially when dealing with high transaction volumes and the need for ultra-low latency. The investment in a modern database technology directly impacted fraud prevention strategies and overall business performance. The importance of a strong database for credit card fraud and other financial fraud applications is clearly demonstrated.
Want AI in your product? Hire the team behind TutorialSearch.io
LLM features, RAG, agents and automation wired into a real product — built by the devs who shipped this site.