Nodal Transaction Graphs for Faster Fraud Ring Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for detecting fraud rings are computationally intensive, require significant processing power, and struggle with managing vast amounts of personally identifiable information (PII) while ensuring real-time detection and minimizing false positives.
Innovation Solution
A nodal data structure is generated from transaction data, collapsing nodes associated with the same entity and using transaction edges to identify potentially fraudulent accounts, reducing the number of edges and comparisons, thereby lowering computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional methods process all transaction data to detect fraud rings, then detection accuracy is improved, but computational resources and processing time increase significantly
Solution Approach 1:
The patent segments the fraud detection process into multiple stages: (1) generating a nodal data structure from transaction data where nodes represent entities and edges represent transactions, (2) identifying suspicious nodes based on fraud indicators, and (3) performing detailed analysis only on connected components containing suspicious nodes. This segmentation allows the system to process only relevant portions of the data, reducing computational resources while maintaining detection accuracy.
Solution Approach 2:
The patent performs preliminary actions by pre-processing transaction data into a nodal data structure and pre-identifying suspicious nodes using fraud indicators before conducting full fraud ring analysis. This preliminary filtering reduces the search space and enables faster subsequent analysis, resolving the contradiction between thorough detection and processing speed.
2Loss of time
If conventional methods analyze all transaction data to ensure real-time detection, then detection timeliness is improved, but processing power requirements increase
Solution Approach 1:
The patent divides the transaction data into disconnected components in the nodal data structure and processes only the connected components containing suspicious nodes. This segmentation dramatically reduces the amount of data requiring real-time processing, enabling timely detection with lower processing power requirements.
Solution Approach 2:
The patent applies partial action by performing comprehensive fraud analysis only on partial subsets of the total transaction data—specifically, only on connected components containing suspicious nodes rather than all transactions. This approach achieves real-time detection capabilities without the need to process the entire dataset, reducing processing power consumption.
3Reliability
If conventional methods store and process all personally identifiable information, then fraud detection completeness is improved, but memory storage costs increase
Solution Approach 1:
The patent segments PII data into the nodal data structure where it is organized by entity relationships rather than stored as a complete dataset. Only PII related to suspicious nodes and their connected components is retained and processed, reducing memory storage requirements while maintaining detection completeness for fraud-related information.
Solution Approach 2:
The patent extracts only the necessary PII data required for fraud detection by identifying suspicious nodes and extracting PII from their connected components. This extraction approach removes unnecessary PII from storage and processing, reducing memory costs while preserving detection completeness for fraudulent activities.
4Measurement precision
If conventional methods compare all accounts to reduce false positives, then accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent segments the comparison process by organizing accounts into disconnected components in the nodal data structure and limiting comparisons to within each component containing suspicious nodes. This segmentation reduces the overall complexity of the system while maintaining precision by focusing comparisons on relevant subsets of accounts.
Solution Approach 2:
The patent applies local quality by performing detailed comparisons and fraud analysis only in local regions of the data—specifically within connected components containing suspicious nodes—rather than applying uniform processing across all accounts. This localized approach reduces system complexity while maintaining high precision in identifying false positives.
Data Source
AI summary
A system includes one or more processors to generate a node graph; determine a first node of the node graph comprises a fraudulent flag indicating a first entity of the first node facilitated a fraudulent transaction; responsive to the determination, identify a set of nodes of the node graph responsive to each node of the set of nodes having a direct transaction edge connection with the first node within the node graph or an indirect transaction edge connection with the first node via at least one node directly connected with the first node; generate a fraudulent flag in a subset of the set of nodes responsive to determining each node in the subset satisfies a matching policy; and generate a record identifying the first node and each node of the subset of nodes.