Ontology Alignment via Semantic Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ontology matching techniques primarily rely on syntactic approaches, lacking effective methods to derive semantically meaningful correspondences between terms in ontologies, which are essential for coherent alignment considering the formal semantics of ontologies.
Innovation Solution
The ASMOV process performs similarity calculations using weighted averages of lexical, structural, and extensional similarities, followed by semantic verification to establish a verified alignment, ensuring that correspondences adhere to the formal semantics of the ontologies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If syntactic approaches are used for ontology matching, then the matching process is simple and fast, but the correspondences lack semantic meaning and coherence
Solution Approach 1:
The ontology matching process is segmented into three distinct phases: (1) pre-alignment extraction using syntactic similarity calculations, (2) semantic verification of correspondences against ontology axioms, and (3) final alignment construction. This segmentation allows the system to first quickly identify candidate correspondences using simple syntactic methods, then systematically verify their semantic validity, thereby improving semantic accuracy without overwhelming complexity
Solution Approach 2:
The system performs preliminary syntactic similarity calculations and pre-alignment extraction before conducting semantic verification. By pre-identifying candidate correspondences based on lexical and structural similarity, the system reduces the search space for semantic verification, making the overall process more efficient while ensuring thorough semantic checking of relevant candidates
2Measurement precision
If multiple similarity measurements are performed with different weights, then the alignment accuracy improves, but the computational complexity increases
Solution Approach 1:
The system calculates multiple types of similarity (lexical, structural, extensional) for all term pairs in the pre-alignment phase, which is computationally intensive. However, semantic verification is then applied to filter and validate these correspondences, ensuring that the high computational investment yields semantically accurate results. The weighted combination of different similarity measurements allows nuanced accuracy improvement
3Reliability
If semantic verification is performed on all correspondences, then the coherence with ontology semantics improves, but the processing time increases
Solution Approach 1:
Semantic verification is applied selectively to correspondences in the pre-alignment rather than uniformly to all possible term pairs. The system focuses verification resources on candidate correspondences that have already demonstrated syntactic similarity, applying semantic checking locally where it is most needed. This approach maintains high semantic coherence for the final alignment while reducing overall processing time
Data Source
AI summary
The present invention relates to computer implemented methods and system for determining correspondences between terms in two or more ontologies. The methods and systems are designed to accept as inputs ontologies in Web Ontology Language (OWL) syntax or any other ontology syntax, to calculate a similarity measure between terms in the ontologies, extract an alignment based on this similarity measure, and verify this alignment according to the semantics contained in the ontologies. This process is designed to be executed iteratively until the similarity measures converge, or until another suitable finalization condition is met. The result of these methods and of the systems implementing these methods is an alignment between two or more ontologies establishing semantic correspondences between the terms in the ontologies.