Fraud Detection via Account Range and Score Stripe Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fraud detection systems in payment card networks are inadequate in detecting widespread fraud attacks, particularly when a set of payment card account data associated with a specific issuer or geographic region is compromised, and they often face challenges with increased network load and computational burdens.
Innovation Solution
A computing system and method that continuously receive and process scored transaction authorization requests, sorting them into account ranges and fraud score range stripes, calculating cumulative metrics over time periods, and detecting fraud events in near real-time by analyzing ratio striping values to identify patterns of fraudulent activity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If payment card transactions are monitored one at a time using traditional fraud detection systems, then individual transaction fraud detection is performed, but widespread fraud attacks on specific issuers or geographic regions cannot be detected
Solution Approach 1:
The system segments transactions by organizing them into account ranges (grouping account numbers) and further into fraud score range stripes (grouping by fraud score). This segmentation enables the system to monitor patterns across groups of transactions rather than individually, allowing detection of widespread fraud attacks on specific issuers or geographic regions while maintaining the ability to process individual transactions.
Solution Approach 2:
The system adds temporal dimension analysis by calculating cumulative metrics over multiple time periods (e.g., current period vs. previous period). This dimensional change from single-point-in-time monitoring to multi-period analysis enables detection of fraud patterns that evolve over time, such as gradual data compromises or coordinated attack waves, without overwhelming complexity.
2Productivity
If traditional fraud detection systems process transactions individually, then computational resources are consumed, but network load increases and processing speed decreases during widespread fraud attacks
Solution Approach 1:
By segmenting transactions into account ranges and fraud score range stripes, the system can process and analyze groups of transactions simultaneously rather than individually. This segmentation enables parallel processing and reduces the computational burden on individual transaction processing, thereby maintaining processing speed even as network traffic volume increases during widespread fraud attacks.
Solution Approach 2:
The system performs preliminary organization of transactions into account ranges and fraud score range stripes before full analysis. This preliminary action groups similar transactions together, enabling more efficient processing and allowing the system to handle higher volumes of network traffic without proportionally increasing computational resources required for each individual transaction.
3Reliability
If fraud detection systems monitor all transactions in detail, then comprehensive fraud detection is achieved, but computational burden and network load increase
Solution Approach 1:
The system applies different levels of monitoring and analysis to different groups of transactions based on their characteristics. By organizing transactions into account ranges and fraud score range stripes, the system can concentrate computational resources on high-risk groups (those with higher fraud scores or suspicious patterns) while using lighter processing for low-risk groups, thereby maintaining detection accuracy without proportionally increasing overall computational burden.
Solution Approach 2:
Segmenting transactions into manageable groups (account ranges and fraud score range stripes) allows the system to process and analyze only the necessary portions of data in detail. This segmentation enables comprehensive fraud detection across all transactions while reducing total computational resources by avoiding redundant analysis of low-risk transactions and enabling parallel processing of high-risk groups.
Data Source
AI summary
A computing system for detecting a pattern of fraudulent network events in a payment card network is configured to continuously receive a plurality of scored transaction authorization requests each including a respective account number and a respective fraud score. The computing system is also configured to sort the scored transaction authorization requests into account ranges based the account number, and sort the transaction authorization requests within each of the account ranges into a fraud score range stripes based on the corresponding fraud score. The computing system is further configured to calculate, for the scored transaction authorization requests within each fraud score range stripe, a ratio of a cumulative metric for a shorter time period over a longer time period, and detect, in near real-time, a fraud event associated with one of the account ranges based on the ratio for one of the fraud score range stripes within the account range.