Fraud Detection Network Graph Cycle Highlighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional network graph analysis for fraud detection becomes cumbersome when dealing with a large number of entities and complex relationships, making it difficult to automatically detect and highlight cycles indicative of potential fraud.
Innovation Solution
A computer-implemented fraud management system that generates a network graph, detects cycles using breadth-first search and depth-first traversal algorithms, and highlights these cycles for further investigation, incorporating features like node and edge type attributes and views to dynamically represent entities and relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If network graph analysis is performed to detect fraud, then fraud detection capability is improved, but analysis complexity increases when dealing with large numbers of entities and complex relationships
Solution Approach 1:
The patent segments the complex network graph analysis into distinct phases: cycle detection phase and fraud analysis phase. The system first identifies all cycles in the network graph using graph traversal algorithms, then performs fraud detection specifically on entities involved in these cycles. This segmentation reduces the overall complexity by focusing computational resources on relevant subgraphs rather than analyzing the entire network.
Solution Approach 2:
The patent extracts and highlights only the cycles from the larger network graph, separating them from the rest of the graph structure. By using visual highlighting techniques to emphasize cyclic paths, the system allows analysts to focus on the extracted fraudulent patterns without being overwhelmed by the complete network complexity.
2Measurement precision
If manual analysis of network graphs is performed, then detailed fraud investigation is possible, but time consumption increases significantly
Solution Approach 1:
The patent performs preliminary cycle detection and highlights fraudulent patterns in the network graph before detailed fraud investigation begins. By pre-identifying and marking cyclic relationships that indicate potential fraud, the system prepares the analysis in advance, allowing investigators to immediately focus on pre-identified suspicious entities rather than manually searching through the entire network.
Solution Approach 2:
The system performs automatic cycle detection and highlighting without requiring manual intervention. The graph traversal algorithms autonomously identify cyclic patterns and the system automatically generates visual highlights, enabling the fraud detection process to serve itself rather than relying on time-consuming manual analysis.
3Reliability
If the entire network graph is analyzed, then comprehensive fraud detection is achieved, but computational resources are wasted on non-fraudulent entities
Solution Approach 1:
The patent applies different analysis quality levels to different parts of the network graph. Cyclic entities receive comprehensive fraud analysis while non-cyclic entities are excluded from detailed investigation. This local quality approach ensures that computational resources are concentrated on entities with fraudulent characteristics (those in cycles) rather than uniformly analyzing all entities in the network.
Solution Approach 2:
The patent performs partial analysis by focusing only on entities involved in cycles rather than analyzing the entire network graph. This selective approach applies the necessary fraud detection action only to the subset of entities that exhibit fraudulent patterns, avoiding excessive computational expenditure on entities that are unlikely to be fraudulent.
Data Source
AI summary
Various embodiments of systems and methods for fraud detection using network analysis are described herein. In an aspect, the method includes receiving a command for generating a network graph for an entity to be investigated for potential fraud. The network graph starts from the entity under investigation and branches outwards displaying other related entities. The entities are represented as nodes with the entity under investigation as an origin node and a relationship between the entities are represented as edges. Once the network graph is generated, a cycle detecting algorithm is executed to detect and mark cycles within the generated network graph. The marked cycles are highlighted to indicate occurrence of potential fraud.