Distributed Transaction Tracing for Multi-Bank Fraud Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to trace funds transferred through multiple bank accounts quickly and efficiently due to the complexity and speed of fraudulent transactions, making it difficult to identify and close associated accounts.
Innovation Solution
A method and apparatus that utilize a network of nodes to trace fraudulent transactions by analyzing account relationships and transaction patterns, enabling rapid identification of fraudulent accounts using a breadth-first traversal and graph theory, and building dispersion trees to track funds in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual investigation is performed using limited bank data views, then data processing simplicity is maintained, but tracing speed and effectiveness deteriorate significantly
Solution Approach 1:
The system segments the banking network into multiple independent nodes (banks), each maintaining its own data. The tracing function is segmented across these nodes through distributed message passing, allowing parallel processing of tracing operations across different banks without requiring a centralized complex system.
Solution Approach 2:
A centralized tracing service acts as an intermediary between banks and investigators. This service receives tracing requests, coordinates the distributed tracing across multiple banks, and aggregates results, simplifying the complexity of coordinating manual investigations across the entire network.
2Reliability
If funds are transferred quickly through multiple accounts, then fraud detection capability is improved, but tracing difficulty increases due to fund mixing and dispersion
Solution Approach 1:
The system performs preliminary actions by pre-establishing account relationship maps and transaction graphs at each bank before fraud occurs. When fraud is detected, these pre-built structures enable rapid tracing through stored relationships rather than real-time analysis of mixed funds, maintaining detection capability while reducing tracing difficulty.
Solution Approach 2:
Instead of attempting to trace physical funds through mixed accounts, the system creates a virtual copy of the transaction flow through message passing. Each bank sends copies of relevant transaction messages to the tracing service, which reconstructs the complete fund trail without physically following the money through mixed accounts.
3Ease of operation
If a centralized database is used to store all transaction data, then data accessibility is improved, but system security and privacy deteriorate
Solution Approach 1:
The system segments transaction data storage across multiple independent banks rather than using a centralized database. Each bank maintains its own local data, reducing the impact of any single point of failure and improving security. The tracing function is enabled through distributed message passing that queries relevant data from multiple segments without requiring central access.
Solution Approach 2:
The tracing service acts as an intermediary that coordinates data access across distributed banks. It receives tracing requests, determines which banks need to be queried, and aggregates results without requiring direct access to centralized databases, thereby maintaining security and privacy while enabling data accessibility for investigation.
Data Source
AI summary
A method of tracing messages through a network of nodes is provided, the method comprising receiving message information corresponding to a first outbound message, the message information comprising a first source identifier and a first destination identifier and determining whether the first source identifier is associated with a set of messages in a storage unit, whereby when the first source identifier is associated with a set of messages, the method comprises producing a trace request, the trace request comprising the first destination identifier and an identifier identifying the set of messages associated with the first source identifier.