Entity Link Analysis for Synthetic Identity Fraud Warnings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems are ineffective in preventing unauthorized access to interactive computing environments by synthetic online entities created for fraudulent purposes, particularly those linked to fraud facilitators like credit mules, which exploit the anonymity of the Internet to maintain synthetic identities.
Innovation Solution
A fraud detection computing system analyzes relationships among online entities and their electronic transactions to identify potential fraud facilitators by applying fraud-facilitation flags to primary entities, providing real-time fraud warnings to client systems during transactions, and securely aggregating data from multiple sources to detect synthetic identities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If entity links are generated between primary entity objects and secondary entity objects to identify fraud rings, then the ability to detect coordinated fraud activity is improved, but the complexity of the system increases due to relationship mapping and graph database requirements
Solution Approach 1:
The patent uses an intermediary system (the entity linking and analysis system) that sits between raw account data and fraud detection outcomes. This intermediary automatically generates entity links, identifies primary and secondary entities, and produces fraud indicators, thereby resolving the contradiction by automating the complex relationship mapping process rather than requiring direct manual system complexity management.
Solution Approach 2:
The patent segments the fraud detection system into distinct functional components: entity link generation, primary entity identification, secondary entity identification, relationship analysis, and fraud indicator generation. This segmentation allows each component to handle specific aspects of the complex task independently, improving reliability through specialized processing while managing overall system complexity through modular architecture.
2Measurement precision
If the system monitors and determines rates at which secondary users change on accounts to update fraud flags, then the detection of synthetic online entities is improved, but the computational resources and processing time increase
Solution Approach 1:
The patent implements periodic monitoring of user changes on accounts rather than continuous analysis. The system determines rates at which secondary users change over time intervals, updates fraud flags based on these periodic assessments, and generates fraud indicators at scheduled intervals. This periodic approach maintains measurement precision for detecting synthetic entities while significantly reducing computational resource consumption compared to continuous monitoring.
3Object-affected harmful factors
If fraud warnings are generated and transmitted to client systems based on target consumer identification in secondary entity objects, then consumer protection is improved, but the quantity of information transmitted and processed increases
Solution Approach 1:
The patent extracts only the essential fraud warning information from the complex entity relationship data and transmits only this extracted essential information to client systems. Rather than transmitting the entire graph database or all entity relationships, the system identifies the critical fraud indicator (presence of target consumer in secondary entity objects with fraud flags) and transmits only this specific warning, thereby protecting consumers while minimizing information volume transmitted.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In some aspects, a computing system can generate entity links between a primary entity object identifying a primary entity for multiple accounts and secondary entity objects identifying secondary entities from the accounts. The computing system can determine a rate at which secondary users change on the accounts. The computing system can update, based on the determined rate, the primary entity object to include a fraud-facilitation flag. The computing system can also service a query from a client system regarding a presence of a fraud warning for a target consumer associated with a consumer system that accesses a service provided with the client system. For instance, the computing system can generate a fraud warning based on the target consumer being identified in a secondary entity object associated with the primary entity object having the fraud-facilitation flag. The computing system can transmit the fraud warning to the client system.