Mappings Between Ontologies in NFDI4ING Terminology Service
The HUMAN-DECISION project aims to develop a neuro-symbolic model for generating a CAD-based precedence graph that represents all the possible assembly sequences of tasks to assemble a product [1]. This paper presents a novel service, initiated in the CoyPu project [2], to compute and explore mappings between all pairs of ontologies ingested into the NFDI4ING Terminology Services (TS) [3]. The mapping tool is used to support the construction of the neuro-symbolic model in the generation of a CAD-based precedence graph. The NFDI4ING TS is an open source platform that provides uniform or standardized access to the latest versions of distributed managed ontologies from different domains. These ontologies are grouped in a collection for engineering related ontologies, such as the NFDI4ING collection [4], but they are not aligned, which limits their reusability in the construction of the aforementioned neuro-symbolic model. The Ontology Lookup Service (OLS) [5] provides the Ontology Xref Service (OxO) service for finding mappings between terms from ontologies and vocabularies. BioPortal [6] is a repository that stores biomedical ontologies. It provides a service for exploring mappings between ontologies ingested in the repository [7]. A drawback of both services is that they do not support the computation and exploration of mappings between ontologies ingested in these repositories and external ontologies. Compared to OxO, the service presented in this paper introduces this innovation in the area of computation and exploration of mappings. In addition, it supports the computation of mappings between ontologies stored on a local machine. The implementation of the mapping service follows the backend-frontend pattern (see Figure 1). In a preprocessing step, the backend computes mappings between all pairs of ontologies using the ontology matching tool LogMap [8] and stores the results in a database. Both the backend component and the database run in a Docker container. The following services are accessible through the frontend [9]: • Exploring mappings with tables. • Exploring mappings using graphs. • Compute mappings between an ontology hosted on a version control system and one or more ontologies ingested into NFDI4ING TS. • Users can also access the mappings REST API to automatically exchange information about mappings between ontologies [10]. A user can explore errors in the computed mappings if they occur [11]. To use this feature, the user must provide as an input a link to the ontology hosted in a version control system and select one or more ontologies ingested into the NFDI4ING TS. Optionally, a reasoner can be used in this scenario to check the satisfiability of classes in the union of the external ontology, selected NFDI4ING TS ontologies, and the computed mappings between them.
Institutions
- Technische Informationsbibliothek (TIB) (DE)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2025-08-04
- DOI
- https://doi.org/10.5281/zenodo.16735777
- Primary Topic
- Manufacturing Process and Optimization
- Type
- article
- Field-Weighted Citation Impact
- 0.00