The availability of terminologies is a critical component of research data management. Without terminologies, meaningful descriptions of research data would not be possible, and the reusability of these data and associated information would be compromised. Accordingly, researchers and funding agencies have a vested interest in the availability of sophisticated and stable terminologies.
In the NFDI4Ing Base Services measure Metadata and Terminology Services ( S-3), we are working on a service offering that allows researchers to easily prepare their results for effective and efficient data and knowledge management to enable their comprehensive reuse.
With our service offering, we first enable researchers to create individual and reusable metadata profiles from standardized metadata blocks via an easy-to-use service (Metadata Profile Service). These metadata blocks are based on formal terminologies that can be automatically retrieved and curated via a second service (Terminology Service). One such terminology is the Metadata4Ing ontology, which is also being developed within S-3. Once the profiles are created, a third service (Metadata Hub) enables the automated input of compatible metadata into metadata repositories to make them available to a wide audience. In this paper, we introduce one of these three interacting services in more detail.
The terminology service addresses requirements related to the access, management, and maintenance of terminologies. In particular, three overarching challenges arise (cf. [EA22]):
- interdisciplinary communities have different designations of terms, which can hinder error-free communication within the community itself and beyond. In this respect, terminologies are specific to a particular community.
- the elements of a terminology change dynamically over time. This affects, among other things, their definitions and the links between them. Consequently, work on terminologies is an active, lengthy, and always ongoing process.
- terminologies must be accepted and reused by the intended community in order to achieve stability and to be available over the long term. Therefore, they must be made known and aligned with existing terminologies as much as possible. This is the only way to ensure active maintenance of terminologies beyond, for example, a funding period.
The terminology service is a web-based application that addresses these challenges at different levels. With respect to the above challenges, we want to enable making clear connections between terminologies from different communities visible and available. Our service is designed to process and retrieve terminologies that support the Semantic Web stack. This approach allows us to standardize the formats of available terminologies and make use of the possibility to support so-called mappings. Mappings are technical solutions to indicate that there is a specific connection between the entry of a terminology A and a terminology B. For example, that an element from terminology A has the same meaning as an element from terminology B. In this way, connections can be made between terminologies from different communities that can also be understood and processed by a computer.
The creation of terminologies is a dynamic process. In certain development cycles, there will therefore always be adjustments to a terminology. Revisions of ontologies or new versions can be easily integrated into the system by us. We are also working on solutions to make old versions of a terminology available for queries. Last but not least, work on terminologies should be community-driven. Without the support of as broad a community as possible, the long-term availability (especially curation) of a terminology cannot be guaranteed. It is precisely the long-term work on terminologies that strengthens their robustness and confidence in their use. In this respect, the terminology service serves as a hub that makes available terminologies visible and thus carries them into the communities.
The terminology service is designed for both human users and machine-to-machine communication. Human users of the service are primarily terminology authors, application developers, or other research-related knowledge workers. For them, we provide comprehensive, faceted full-text search and browsing of available terminologies. Once a matching terminology or entry of a terminology is found, it can be visualized, and further metadata can be queried. For example, data curators are already using the service to link metadata in our data repository to matching controlled vocabularies. We are also currently working on integrating a tool for collaborative work on terminologies and many other tools for terminology analysis and visualization.
Machine retrieval of terminologies for integration into data management applications has been technically solved using a REST interface. A wide range of different, highly differentiated information can be queried using this interface. There are already several application examples for the machine query of terminologies, e.g., the aforementioned creation of individual metadata profiles, a browser plugin (TermClick, [TE22]) that queries the terminology service to provide definitions for a word marked on a website, or the enrichment of Knowledge Graphs with definitions.
[EA22] Engel F., Auer S. 2022; Collaborative Work on Ontologies – A Report International Workshop on Data-driven Resilience Research (D2R2). Not yet published ↩