Payment Record Relationship Scoring for Fraud Risk Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to accurately identify and understand relationships between multiple payment cards or payment accounts, particularly when they are shared by different users or used by a single user, complicating fraud detection and risk assessment.
Innovation Solution
A system and method that analyze data fields such as locality, shipping address, email, and device information to identify matches between payment records, generating a relationship score to represent these connections, and provide graphical representations for user interface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If relationship identification between payment records is performed, then fraud detection accuracy is improved, but system complexity increases
Solution Approach 1:
The system segments the relationship identification process into multiple independent matching modules, each handling a specific data field (locality, shipping address, email, device). This segmentation allows the complex task of relationship identification to be broken down into manageable components, improving fraud detection accuracy while keeping each module's complexity low and independent.
Solution Approach 2:
The patent introduces an intermediary matching mechanism that compares data fields between payment records to identify relationships. This intermediary layer of comparison and matching simplifies the overall system architecture by providing a standardized approach to relationship detection, thereby improving fraud detection accuracy without proportionally increasing system complexity.
2Measurement precision
If multiple data fields are compared to identify relationships, then relationship identification accuracy is improved, but processing time increases
Solution Approach 1:
The comparison process is segmented into separate operations for each data field (locality, shipping address, email, device). This allows the system to process each field independently and efficiently, maintaining high relationship identification accuracy while reducing overall processing time through parallelizable operations.
Solution Approach 2:
The system performs partial comparisons by evaluating multiple data fields to different degrees of strictness. Not all fields need to be compared with equal rigor depending on the specific relationship being detected, allowing the system to achieve sufficient accuracy without the time cost of exhaustive comparison of all fields with maximum strictness.
3Reliability
If relationship data is stored and processed, then fraud detection capability is improved, but data management complexity increases
Solution Approach 1:
The data management system segments stored relationship data by type and source, organizing it in a structured manner that improves fraud detection capability while reducing management complexity. Each data field (locality, address, email, device) is managed as a separate entity, simplifying queries, updates, and maintenance operations.
Data Source
AI summary
Examples include a system that identifies relationships between records. Data corresponding to a plurality of records is accessed, wherein the data is related to different consumer cards. Data fields of the data are compared to identify one or more matches between the plurality of records. A plurality of relationships between the plurality of records is determined based at least on the identified one or more matches. A score is generated based on the determined plurality of relationships. The system produces risk assessment data with enhanced reliability and greater accuracy while reducing system resource usage.