Semantic Framework for Digital Entities Using Knowledge Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The development of metaverse-based digital twins faces challenges due to the manual and iterative process of translating domain knowledge into executable code, labor-intensive visualization efforts, and compatibility issues between different simulation technologies and platforms.
Innovation Solution
An automated semantic mechanism that converts controlled natural language data into executable queries, operationalizes entity models using model ontologies, and transforms semantic bindings into knowledge graphs, enabling efficient generation of metaverse-based digital twins.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual and iterative process is used to translate domain knowledge into executable code, then domain expertise can be captured, but the process becomes time-consuming and labour-intensive
Solution Approach 1:
The patent replaces the manual mechanical process of translating domain knowledge into code with an automated semantic mechanism. The system uses natural language processing and semantic parsing to automatically convert domain expert descriptions into executable code, eliminating the need for manual translation while preserving domain knowledge accuracy.
Solution Approach 2:
The system enables domain experts to directly input their knowledge in natural language without requiring interaction with data modelers or software engineers. The automated mechanism processes this input and generates executable code independently, allowing the domain knowledge system to serve itself without external intervention.
2Manufacturing precision
If data modelling requires interaction between data modelers and domain experts, then domain constructs can be accurately modeled, but complexity and time overhead increase
Solution Approach 1:
The patent extracts the data modeling function from the complex interactive process and consolidates it into an automated semantic mechanism. The system directly parses natural language input from domain experts and generates domain models without requiring intermediate data modelers, thereby reducing process complexity while maintaining modeling accuracy.
Solution Approach 2:
The automated semantic mechanism acts as an intermediary between domain experts and the digital twin system. It translates natural language descriptions directly into structured domain models and executable code, eliminating the need for human data modelers as intermediaries and simplifying the overall process.
3Adaptability or versatility
If manual translation of domain knowledge into executable code is performed, then customization to domain rules is achieved, but substantial labour effort is required
Solution Approach 1:
The patent replaces manual translation operations with an automated semantic parsing system that converts natural language domain knowledge into executable code. This substitution maintains the ability to customize to domain-specific rules while dramatically improving development efficiency by eliminating manual labour.
Solution Approach 2:
The system performs preliminary semantic analysis and code generation based on natural language input before deployment. By pre-processing domain knowledge into structured formats and executable code in advance, the system enables rapid customization without requiring substantial labour effort during the deployment phase.
4Manufacturing precision
If digital twins are constructed in a siloed manner, then domain-specific accuracy is maintained, but integration of multiple digital twins becomes complex
Solution Approach 1:
The patent implements a universal semantic framework that can process domain knowledge from multiple domains using the same automated mechanism. The system maintains domain-specific accuracy through semantic parsing while providing a standardized interface for integration, thereby reducing integration complexity without sacrificing domain fidelity.
Solution Approach 2:
The system segments the digital twin construction process into independent domain-specific modules, each processed by the automated semantic mechanism. This segmentation allows each digital twin to maintain its domain-specific accuracy while the standardized output format from the semantic mechanism simplifies integration between multiple digital twins.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
An apparatus and method (200) for defining a semantic framework for a plurality of entities in a computer simulated environment. The method (200) includes receiving data from one or more data acquisition devices (104), translating, the data into one or more executable queries to extract data relating to the one or more elements from the knowledge database (106). The method (200) further includes assembling an entity model with one or more elements identified from the CNL data, each element operationalized using a model ontology and instantiating each element in the entity model using a plurality of instances for each element in the entity model to generate a semantic binding among the plurality of elements. Furthermore, the method (200) includes transforming the semantic binding into a knowledge graph using standard graph databases and translating the semantic description from the knowledge graph into an executable configuration for the metaverse.