Graph-Based Synthetic Entity Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cybersecurity measures fail to effectively detect synthetic online entities, which can be used for fraudulent activities due to their ability to create deep, fake identities that evade detection in electronic transactions, exploiting the anonymity of the Internet.
Innovation Solution
A computing system generates a data structure with nodes and links representing online entities and their associations, analyzing connectivity to identify potential synthetic entities by exceeding a threshold connectivity, thereby alerting for potentially fraudulent activity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional automated computing systems are used to service loan applications, then processing efficiency is maintained, but synthetic entities can fraudulently apply for loans without detection due to their ability to create fake identities
Solution Approach 1:
The patent transitions from traditional single-entity verification to multi-dimensional network analysis by constructing graphs that map relationships among entities, devices, and locations. This dimensional expansion enables detection of synthetic entities through their abnormal connectivity patterns rather than relying solely on individual entity verification.
Solution Approach 2:
The system introduces an intermediary analytical layer that processes transaction data and generates connectivity metrics before final lending decisions. This intermediary layer analyzes the network relationships and flags suspicious patterns, allowing the core lending system to remain relatively simple while adding detection capability.
2Difficulty of detecting and measuring
If deep fake identities are created for fraudulent transactions, then anonymity is achieved, but unusual interconnectivity patterns are generated that can be detected through graph analysis
Solution Approach 1:
The system implements feedback loops where connectivity metrics are continuously calculated and compared against thresholds. When abnormal patterns are detected, alerts are generated and investigations are initiated, creating a feedback mechanism that improves detection over time while maintaining operational efficiency.
Solution Approach 2:
The patent replaces traditional mechanical verification methods with automated graph-based analytical systems. Instead of manual identity verification, the system uses automated algorithms to analyze connectivity patterns, reducing information loss while making detection more difficult for fraudsters.
3Productivity
If multiple synthetic entities are used to conduct clandestine activities, then fraudulent transactions increase, but the degree of connectivity among these entities exceeds normal thresholds enabling detection
Solution Approach 1:
The system changes the parameters used for evaluating transaction legitimacy from individual entity characteristics to collective connectivity metrics. By monitoring the degree of connectivity among entities and comparing against dynamically adjusted thresholds, the system can identify fraudulent activity patterns while allowing normal transaction volumes.
Data Source
AI summary
Examples are disclosed for detecting synthetic online entities that may be used for fraudulent purposes or other purposes. In some aspects, a computing system can generate a data structure that includes nodes and links between the nodes. The nodes can represent online entities and the links can represent geographic associations or transactional associations between pairs of online entities. These associations can be identified from electronic transactions involving the online entities. The computing system can determine, from the links between the nodes, that a degree of connectivity among a subset of the nodes exceeds a threshold connectivity. The degree of connectivity indicates electronic communications involving online entities represented by the subset of the nodes. The computing system can transmit, based on the degree of connectivity exceeding the threshold connectivity, an alert indicating a potential synthetic entity (e.g., potentially fraudulent activity) within the subset of the nodes.