Semantic Engine for BI Metadata and OWL Model Query Translation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Business Intelligence (BI) applications face limitations in querying and integrating data from semantic models, such as Web Ontology Language (OWL) models, due to the lack of effective mapping and integration with traditional BI metadata models, which hinders comprehensive data analysis and reporting.
Innovation Solution
A computer-implemented method and system that constructs BI metadata models from semantic models, allowing queries to be executed across multiple data sources, combining semantic and data source result sets to generate combined reports, using specialized mapping systems and engines to translate queries between BI and OWL formats.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If BI applications use traditional BI metadata models for querying, then querying traditional data sources is efficient, but querying semantic models (OWL) is not supported
Solution Approach 1:
The patent introduces a semantic engine as an intermediary component between BI applications and semantic models. This engine translates BI metadata model queries into OWL-compatible query formats, enabling BI applications to query semantic models without direct integration complexity. The semantic engine acts as the mediator that handles the translation and communication protocols between the two different data model paradigms.
Solution Approach 2:
The patent creates a unified query execution framework that can handle both traditional data sources and semantic models through a common interface. The BI metadata model is extended to accommodate semantic model querying capabilities while maintaining backward compatibility with traditional data source queries, making the system universally applicable to multiple data types without requiring separate specialized systems.
2Loss of information
If BI applications integrate multiple data sources, then comprehensive data analysis is achieved, but integration with semantic models is hindered by lack of mapping
Solution Approach 1:
The patent segments the integration process into distinct functional layers: a translation layer that converts between BI metadata model and OWL ontology formats, and an execution layer that handles query processing. This segmentation allows each layer to specialize in its specific function, making the overall integration process more manageable and easier to implement without requiring complete re-engineering of the BI system.
Solution Approach 2:
The patent transforms queries by changing their parameter representation format - converting BI metadata model parameters into OWL ontology-compatible parameters and vice versa. This parameter transformation approach enables seamless integration by adjusting the data representation format rather than requiring fundamental changes to the query structure or execution logic.
3Loss of information
If semantic models are used for deep data discovery, then reasoning capabilities are enhanced, but integration with BI reporting is limited
Solution Approach 1:
The semantic engine serves as a mediator that bridges semantic models and BI reporting systems. It translates semantic query results into formats suitable for BI reporting while preserving the reasoning capabilities and deep data discovery insights provided by the semantic models. This intermediary layer enables the reporting system to leverage semantic reasoning without requiring direct semantic model integration.
Data Source
AI summary
Provided are a computer implemented method, computer program product, and system for generating a combined report. One or more queries are constructed using a Business Intelligence (BI) metadata model and one or more query declarations. The one or more queries are executed on a semantic model to obtain one or more semantic result sets. One or more data source result sets are received from one or more data sources. The one or more semantic result sets and the one or more data source result sets are combined to construct one or more combined result sets. The one or more combined result sets are rendered to form a combined report.