Knowledge Graph Entity Generation via Causality Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies in knowledge graph approaches are limited to link prediction and entity embedding, and do not provide for entity generation, which is essential for adding new entities (nodes) in a graph to meet specific requirements.
Innovation Solution
A computer-implemented method that detects counterfactual causes in a causality graph connected to a knowledge graph, generates new entities in the knowledge graph by embedding them in a latent space relative to existing entities, and simulates the change of causes resulting from the new entity generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing knowledge graph approaches are used for link prediction and entity embedding, then the system can perform basic graph operations, but the system cannot generate new entities (nodes) in the graph
Solution Approach 1:
The system segments the entity generation process into distinct components: counterfactual cause detection in the causality graph, latent space embedding generation, and new entity creation in the knowledge graph. This modular approach enables entity generation capability while managing system complexity through divided functional blocks.
Solution Approach 2:
The patent introduces a causality graph as an intermediary structure between existing knowledge graphs and new entity generation. The causality graph detects counterfactual causes that guide the generation of new entities, serving as a mediator that bridges the gap between existing graph operations and novel entity creation capabilities.
2Reliability
If new entities are generated in the knowledge graph, then the target property can be improved, but the system requires counterfactual cause detection and simulation capabilities
Solution Approach 1:
The system performs preliminary actions by detecting counterfactual causes in the causality graph before generating new entities in the knowledge graph. This preliminary cause detection and simulation step ensures that new entities are generated with purpose and direction, improving target properties while managing complexity through structured preprocessing.
Solution Approach 2:
The patent implements feedback mechanisms where the simulation of cause changes resulting from new entity generation feeds back into the system. This feedback loop allows the system to evaluate the impact of generated entities and refine future generation processes, improving reliability through iterative optimization.
Data Source
AI summary
A computer-implemented, machine learning method for incorporating a new entity in a knowledge graph for optimizing or improving a target property includes detecting counterfactual causes in a causality graph that are to be modified to achieve the target property. The causality graph is connected to the knowledge graph by links representing semantic relations. The new entity is generated in the knowledge graph by embedding the new entity in a latent space of the knowledge graph relative to existing entities. A change of causes in the causality graph resulting from generating the new entity in the knowledge graph is simulated. The method can be applied, for example, to use cases in medical/healthcare, smart cities or smart agriculture, for example, to support decision making using Artificial Intelligence (AI).