Graph Database Topology Analysis for Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional adaptive authentication systems underestimate risk scores due to focusing solely on individual user behavior, neglecting complex relationships between users and devices, leading to impractical computational costs and inadequate fraud detection.
Innovation Solution
Generating a relational graph from historical transaction data to represent connections between users and transaction devices, using a graph database to analyze topologies and compute risk scores, allowing for a more accurate assessment of transaction risk.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional adaptive authentication systems focus solely on individual user behavior profiles, then the system is simple to implement and compute, but the fraud detection accuracy deteriorates because complex relationships between users and devices are neglected
Solution Approach 1:
The patent transitions from analyzing single-user behavior profiles to constructing multi-dimensional relational graphs that capture connections between users, devices, and transactions. This dimensional expansion enables detection of fraud patterns that span multiple users and devices, significantly improving fraud detection accuracy while the graph-based approach provides a structured framework to manage the increased complexity
2Measurement precision
If historical records are maintained for every user, device or payee to assess fraud risk, then fraud detection capability improves, but computational costs become impractically high
Solution Approach 1:
The patent extracts and analyzes only the relevant relational structures from historical data by constructing graphs that focus on connections between entities involved in transactions. This selective extraction of relational patterns rather than processing all historical records reduces computational costs while maintaining fraud detection capability
Solution Approach 2:
The system pre-computes and stores relational graphs representing connections between users, devices, and transactions. These pre-computed graph structures enable efficient fraud risk assessment by allowing the system to quickly query pre-established relationships rather than computing them in real-time during transaction processing
3Adaptability or versatility
If traditional relational database models are used to store transaction histories, then data storage is simple, but the ability to represent and query complex relationships between users and devices deteriorates
Solution Approach 1:
The patent introduces graph database models that add a new dimensional layer for representing relationships, moving beyond traditional tabular relational databases. This graph-based dimension enables natural representation of complex multi-entity relationships while the underlying graph database technology provides efficient query capabilities for traversing these relationships
Data Source
AI summary
An improved technique involves generating, from historical transaction data, a relational graph that represents connections between users who initiate transactions and transaction devices used to carry out the transactions. By supplementing traditional relational database models with a tool such as a graph database, a risk analysis server is able to express users and transaction devices as nodes in a graph and the connections between them as edges in the graph. The risk analysis server may then match the topology of the graph in a neighborhood of the user initiating the transaction to a known topology that is linked to an indication of risk. In some arrangements, this topology is an input into a risk model used to compute a risk score for adaptive authentication.