Ontology-Based Decision Intelligence With Automated Data Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional decision intelligence systems fail to provide satisfactory solutions for enterprise and Industrial Internet of Things (IIoT) ecosystems due to the lack of context in machine learning algorithms, leading to ineffective decision-making processes.
Innovation Solution
A decision intelligence system that allows data experts to input business questions and generate a semantic ontology-based business knowledge model, which is mapped to data sources and used to create APIs for generating views of business data, leveraging a platform application with a studio application to build models and perform question execution and data mapping.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If machine learning algorithms are used in conventional decision intelligence systems, then automated decision-making capability is improved, but the lack of contextual understanding (asset supply chain, engineering, commercialization, service) causes decision accuracy to deteriorate
Solution Approach 1:
The patent introduces a semantic layer as an intermediary between raw data and machine learning algorithms. This semantic layer contains structured knowledge models with ontologies that provide contextual understanding of asset supply chain, engineering, commercialization, and service domains. The semantic layer translates and enriches raw data with contextual meaning before feeding it to ML algorithms, thereby maintaining automation while improving decision accuracy through domain-specific context.
Solution Approach 2:
The system performs preliminary action by pre-building comprehensive knowledge models and ontologies that encode domain expertise and contextual relationships before the decision-making process. These pre-structured knowledge models include asset hierarchies, supply chain relationships, engineering specifications, and service workflows. When ML algorithms need to make decisions, they can directly query these pre-prepared contextual models rather than learning context from raw data, improving both speed and accuracy.
2Productivity
If visualizations and analytical tools are built upon automatically created models, then system development speed is improved, but the models fail to capture domain expertise and business context causing decision quality to deteriorate
Solution Approach 1:
The patent implements self-service by enabling domain experts to directly author and refine knowledge models using intuitive visualization tools. The system allows experts to drag-and-drop concepts, define relationships, and encode domain knowledge without requiring programming expertise. This self-service capability ensures that the models accurately reflect domain expertise while maintaining rapid development through automated model creation and refinement workflows.
Solution Approach 2:
The system provides universal knowledge models that can serve multiple functions across different decision-making scenarios. The ontologies and knowledge models are designed to be reusable across asset supply chain, engineering, commercialization, and service domains. A single knowledge model can support multiple analytical tools and visualizations, improving decision quality through consistent domain expertise while accelerating development through model reuse.
3Measurement precision
If extensive data mapping is performed to align data sources with business models, then data accuracy is improved, but the time and complexity of system implementation increases
Solution Approach 1:
The semantic layer acts as an intermediary that automatically performs data mapping between diverse data sources and the knowledge model. Instead of manually mapping each data source to the business model, the system uses semantic ontologies to automatically interpret and align data from multiple sources with the appropriate concepts in the knowledge model. This automated semantic mapping maintains data accuracy while dramatically reducing implementation time and complexity.
Solution Approach 2:
The system performs preliminary action by pre-defining standardized ontologies and data schemas that anticipate common data mapping requirements. Knowledge models are pre-structured with expected data formats and relationships, allowing the system to automatically match incoming data sources to these predefined structures. This preliminary preparation eliminates the need for extensive custom mapping during implementation, reducing both time and complexity while maintaining accuracy through validated data schemas.
Data Source
AI summary
A decision intelligence system permits a data expert to input information to define business questions and a business knowledge model. A semantic ontology is generated from the business knowledge model. The business knowledge model is mapped to data sources and used to generate information for generating APIs for generating views of business data, such as machine generated data.