Assertion Management for Ontology Reasoning Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ontology-based reasoning systems face inefficiencies in reasoning performance due to the large number of assertions and axioms, which can hinder effective deduction and increase processing time.
Innovation Solution
An assertion management method and apparatus that collect and process 'SameAs' assertions to create new assertions by substituting individuals, thereby simplifying the assertion format and improving reasoning efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the number of assertions and axioms is increased to improve reasoning completeness, then reasoning coverage is improved, but reasoning performance deteriorates due to increased processing time
Solution Approach 1:
The patent applies preliminary action by performing substitution based on SameAs assertions before the main reasoning process. The assertion management apparatus pre-processes assertions to replace equivalent individuals, so that when reasoning occurs, the assertion set is already optimized. This reduces the time required during actual reasoning while maintaining completeness.
Solution Approach 2:
The patent changes the parameter of assertion representation by substituting individual identifiers based on SameAs relationships. Instead of using original individual identifiers, the system transforms assertions to use canonical or representative identifiers, thereby reducing redundancy and improving reasoning performance without losing any logical information.
2Reliability
If the number of assertions and axioms is increased to improve reasoning completeness, then reasoning coverage is improved, but device complexity increases
Solution Approach 1:
The patent extracts the complexity of assertion management into a separate dedicated apparatus. The assertion management apparatus is isolated from the main reasoning engine, handling substitution and optimization tasks independently. This separation reduces the complexity burden on the overall reasoning system while maintaining complete reasoning capabilities.
Solution Approach 2:
The assertion management apparatus serves as an intermediary between the raw assertion set and the reasoning engine. It mediates by pre-processing assertions through substitution based on SameAs relationships, delivering an optimized assertion set to the reasoner. This intermediary layer simplifies the reasoning engine's workload while preserving reasoning completeness.
3Productivity
If assertion format is simplified to improve processing efficiency, then reasoning performance is improved, but information representation capability deteriorates
Solution Approach 1:
The patent uses copying by creating new assertions with substituted individual identifiers while preserving the original assertion meanings. The substitution based on SameAs relationships creates equivalent representations that maintain all necessary information for reasoning, just in a more efficient format suitable for processing.
Data Source
AI summary
An assertion management method includes collecting a SameAs assertion from a plurality of assertions; searching the plurality of assertions for any assertion having a first individual of the SameAs assertion as an instance; and creating a new assertion having a second individual of the SameAs assertion as an instance from each found assertion.