Payment Network Route Analysis for Neural Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The decentralized nature of electronic payment networks makes it difficult to detect fraud and identify responsible parties, as users and systems involved in payments can commit fraud, and no entity has a complete view of the network topology.
Innovation Solution
A neural network is trained using payment records to determine fraud probabilities by analyzing payment data, including user identities and network routes, and generates fraud probabilities for users, ledgers, and connectors involved in the payment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the electronic payment network is decentralized to allow users to send and receive payments, then the network's accessibility and flexibility are improved, but the ability to detect fraud and identify responsible parties deteriorates
Solution Approach 1:
The patent introduces a centralized fraud detection system that acts as an intermediary between decentralized payment participants. This system receives payment data from various nodes in the decentralized network, processes it through machine learning models, and generates fraud probabilities. The intermediary consolidates information that would otherwise be scattered across the decentralized network, enabling effective fraud detection without requiring changes to the underlying decentralized architecture.
Solution Approach 2:
The system implements feedback mechanisms where fraud detection results are fed back into the network to improve future detection accuracy. The machine learning models are trained on historical payment data and fraud patterns, continuously refining their ability to detect fraudulent activities. This feedback loop allows the system to adapt to new fraud techniques while maintaining the decentralized nature of the payment network.
2Measurement precision
If payment data is processed through a centralized fraud detection system, then fraud detection accuracy is improved, but the decentralized nature of the network is compromised
Solution Approach 1:
The fraud detection system is segmented into modular components that can be independently deployed and scaled. The system divides fraud detection into separate analytical functions (e.g., transaction pattern analysis, user behavior analysis, network route analysis) that can be processed independently. This segmentation allows the system to achieve high detection accuracy through specialized processing while maintaining flexibility in deployment architecture and reducing overall system complexity.
Data Source
AI summary
Systems and techniques are provided for electronic payment network security. Payment data including an origin and a destination for a payment in an electronic payment network may be received. A route of the payment in the electronic payment network may be estimated based on the origin and the destination. The estimated route of the payment in the electronic payment network may be input to a neural network. Fraud probabilities may be determined using the neural network. A fraud probability may include a value indicating a probability of fraud in the payment in the electronic payment network.