Entity Relationship Graph Confidence Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing relationship graph applications struggle to clearly represent multi-hop relationships among entities due to overcrowding, often relying on one-hop graphs that provide limited insights.
Innovation Solution
A system generates a graphical representation of entities with source nodes, related nodes, and supplemental nodes, using confidence scores to identify and display only relevant relationships, thereby traversing the graph to produce results efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If a multi-hop relationship graph is generated to show more relationships, then the completeness of relationship information is improved, but the graph becomes too crowded for users to understand
Solution Approach 1:
The relationship graph is segmented by hop distance, with source nodes representing input entities, related nodes representing first-hop relationships, and supplemental nodes representing second-hop relationships. This segmentation allows the system to display multi-hop relationships while maintaining visual organization through distinct node types and confidence score thresholds, preventing overcrowding while preserving relationship completeness.
2Ease of operation
If a one-hop relationship graph is generated to maintain clarity, then user understanding is improved, but the amount of relationship information is limited
Solution Approach 1:
The system dynamically adjusts the relationship graph by generating different views based on confidence scores. High-confidence relationships are displayed as related nodes in the primary view, while additional relationships are available as supplemental nodes. This dynamic approach allows the graph to maintain clarity for common cases while providing access to more complete relationship information when needed.
3Loss of information
If all relationships are displayed in the graph, then relationship completeness is improved, but processing time increases
Solution Approach 1:
The system applies partial action by generating relationship graphs with confidence score thresholds that filter out low-confidence relationships. This allows the system to process and display only the most relevant relationships (those meeting the threshold), reducing processing time while maintaining sufficient relationship completeness for practical purposes. The threshold can be adjusted based on specific needs.
Data Source
AI summary
Methods, systems, and computer program products are provided for processing a request regarding relationships among instances of entities. A graphical representation of instances of entities is generated and includes one or more source nodes, each representing an instance of an input entity of a request, and one or more related nodes, each representing an instance of a second entity related to one or more corresponding instances of the input entity and associated with a corresponding confidence score for the relationship. Each of the one or more related nodes associated with a confidence score satisfying a threshold is identified. One or more supplemental nodes are added to the graphical representation, each of which represents a corresponding instance of a third entity with a relationship to a corresponding instance of the second entity. The graphical representation is traversed to identify relationships between instances of entities and produce results for the request.