Identity Chaining Fraud Detection via Shared Transaction Links
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fraud detection methods in payment card transactions rely heavily on historical data and struggle to effectively identify fraudulent activities across multiple merchants, as they lack the ability to link transactions through shared identities in real-time.
Innovation Solution
The identity chaining method links current transactions to other transactions through shared identities, forming chains that can indicate fraudulent activity by reaching a maximum limit of variations, and utilizes negative lists to flag suspicious transactions, even across different merchants, through a centralized payment processing network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If historical data analysis methods (velocity method and identity morphing) are used for fraud detection, then fraud can be detected based on past transaction patterns, but the system cannot effectively identify fraudulent activities across multiple merchants in real-time
Solution Approach 1:
The system pre-establishes identity chains by linking identities (account numbers, email addresses, IP addresses, device IDs, billing addresses, shipping addresses, phone numbers) across transactions before fraud occurs. This preliminary structuring of identity relationships enables real-time fraud detection without requiring complex historical analysis at the moment of transaction, thus resolving the contradiction between detection accuracy and real-time capability
Solution Approach 2:
The patent introduces identity chains as an intermediary structure that connects transactions across different merchants. Instead of directly analyzing historical data from multiple merchants in real-time, the system uses pre-built identity chains as a mediator to quickly assess fraud risk, enabling cross-merchant fraud detection without the computational burden of real-time historical analysis
2Measurement precision
If identity variations are monitored to detect fraud, then fraudulent transactions can be identified when a maximum limit of variations is reached, but the system lacks the ability to link transactions through shared identities across different merchants
Solution Approach 1:
The identity chain structure serves multiple functions: it tracks identity variations within a single merchant's transactions and simultaneously links transactions across different merchants through shared identities. This multi-functional design enables the system to both measure identity variations precisely and adapt to cross-merchant fraud patterns, resolving the contradiction between measurement precision and adaptability
3Reliability
If negative lists are maintained for each merchant to flag fraudulent identities, then suspicious transactions can be identified, but merchants cannot detect fraud patterns that span across multiple merchants
Solution Approach 1:
The patent merges individual merchant negative lists into a centralized identity chain database that aggregates fraud information across all merchants. By combining the information from multiple merchants' negative lists while maintaining the ability to trace identity relationships, the system enables both suspicious transaction identification and cross-merchant fraud pattern detection, resolving the contradiction between reliability and information loss
Data Source
AI summary
An identity chaining fraud detection method that allows each current transaction to be linked to other transactions through commonly shared identities. Over a period of time the links create a chain of associated transactions which can be analyzed to determine if identity variances occur, which indicates that fraud is detected. Additionally, if a specific identity is detected as being fraudulent, that identity can be tagged as fraudulent and can be referenced by a plurality of other merchant transaction chains to determine fraud.