Hybrid Graph and Label Model for Entity Association
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed computing environments, identifying and tracking associations between entities related to credit fraud is inefficient due to the exponential growth of relationships in graph models, making it difficult to detect fraud effectively.
Innovation Solution
Implementing a hybrid approach that combines graph models and label models, where entities are optimized by associating related entities with labels, reducing the parsing order from exponential to linear, and using a pointer mechanism to connect records, allowing for efficient association and optimization of entities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If graph models are used to represent entity relationships, then comprehensive entity association tracking is achieved, but parsing complexity grows exponentially
Solution Approach 1:
The patent segments the complex graph model into two distinct components: a graph model for storing entity relationships and a label model for storing optimized entity labels. This segmentation allows the system to maintain comprehensive relationship tracking while reducing parsing complexity by directing queries to the appropriate model based on optimization status.
Solution Approach 2:
The patent introduces an optimization status flag as an intermediary mechanism that mediates between the graph model and label model. This flag indicates whether an entity has been optimized, allowing the system to efficiently route queries without traversing the entire graph structure, thus reducing parsing complexity while maintaining association tracking.
2Reliability
If graph models are used to represent entity relationships, then complete relationship mapping is achieved, but resource consumption increases
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing optimized labels for entities in the label model during idle periods or when changes occur in the graph model. This allows query operations to retrieve pre-computed labels directly without performing expensive graph traversals at query time, significantly reducing resource consumption while maintaining complete relationship mapping capability.
3Productivity
If entity optimization is implemented using label models, then parsing efficiency improves, but system complexity increases
Solution Approach 1:
The patent implements dynamics by making the system adaptive through the optimization status flag. The system dynamically adjusts its behavior based on whether an entity is optimized: for optimized entities, it queries the simple label model for fast parsing; for unoptimized entities, it queries the comprehensive graph model. This dynamic approach improves parsing efficiency for common cases while maintaining the ability to handle complex relationships when needed.
Data Source
AI summary
Systems and methods for efficient association of related entities. The method may comprise accessing a database of records, using a processor, to identify a set of unoptimized entities represented by one or more nodes in a graph model, a connection between a first node and a second node in the one or more nodes representing an association between a first entity represented by the first node and a second entity represented by the second node; determining the first entity is unoptimized; and determining a set of related entities for the unoptimized first entity in the graph model, the graph model having at least one common entity with a corresponding label model.