Asynchronous Graph Computing for Electronic Communication Risk Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic communication processing systems face inefficiencies in data retrieval and risk analysis due to synchronous processing limitations, leading to poor user experience and decreased catch rates of malicious activities, particularly in high-volume transaction environments with strict service level agreements (SLAs).
Innovation Solution
Implementing asynchronous graph computing techniques that perform data retrieval and risk analysis prior to transaction initiation, utilizing a graph database system to execute multi-hop queries and store results in a non-relational database, enabling complex computations and efficient data updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If synchronous processing is used for risk analysis, then processing simplicity is maintained, but detection speed and user experience deteriorate due to strict SLA constraints
Solution Approach 1:
The system performs risk analysis computations asynchronously before transactions are fully initiated. Graph computations are triggered by events such as account creation or login, and results are cached in advance. When a transaction occurs, the pre-computed results are immediately retrieved, enabling fast detection without blocking the transaction flow and maintaining simplicity at the point of use.
2Measurement precision
If complex graph computations are performed, then detection accuracy improves, but resource consumption increases
Solution Approach 1:
Complex graph computations are performed in advance asynchronously and results are cached. This shifts the computational burden to off-peak times when resources are more available, while transactions benefit from immediate retrieval of pre-computed results, reducing real-time resource consumption.
Solution Approach 2:
The system computes graph data locally and specifically for each account or entity rather than performing global computations. This targeted approach maintains high detection accuracy by analyzing relevant connections while significantly reducing overall resource consumption by avoiding unnecessary computations.
3Reliability
If more graph nodes are accessed for analysis, then detection completeness improves, but processing time exceeds SLA requirements
Solution Approach 1:
The system pre-computes and caches graph analysis results for multiple hops of connections before transactions occur. This allows comprehensive analysis of extended networks (5+ hops) to be completed in advance, ensuring detection completeness while enabling rapid retrieval that meets SLA requirements during actual transactions.
Data Source
AI summary
Techniques are disclosed for detecting risk via a server system that receives, based on user activity at a device, an indication of a trigger event corresponding to a potential electronic communication. After receiving the trigger event indication and prior to receiving an indication of initiation of the electronic communication, the system executes a set of computations for the communication, including performing a multi-hop query to a graph database storing a graphical representation of a plurality of communications and storing results of the execution. In response to initiation of the electronic communication, the system retrieves, using information corresponding to entities involved in the electronic communication, one or more portions of the results of executing the set of computations the database. The system determines, based on the retrieved results, whether to approve the electronic communication. The disclosed techniques may advantageously decrease unsecure electronic communications relative to real-time risk detection techniques.