Ontologies are powerful tools for structuring and formalizing knowledge and enabling interoperability. They play a crucial in research data management by making data machine-interpretable and facilitating the integration of diverse datasets from various researchers. Ideally, the integrated datasets contribute to global use cases aimed at inferring new knowledge. To achieve this, it is essential that researchers align the data collected for their local use cases with existing domain ontologies. However, the data needs for local use cases might not readily fit into these domain ontologies without further adaptation. As a consequence, researchers often refrain from reusing the ontologies and collect their local data on an ad hoc basis using improvised schemas.

We surveyed current approaches to ontology-based data acquisition, revealing that while some methods exist for generating forms from ontologies, there are currently no significant approaches that enable researchers to customize these ontologies to fit their individual needs.

To address this gap, we developed a prototype aimed at exploring the challenges of adapting ontologies for specific use cases. Our form-based approach empowers researchers to reuse ontologies without requiring substantial expertise in ontology engineering.

This prototype allows researchers to define ontology-based data collection forms tailored to their specific needs, ultimately generating ontology-compliant knowledge graphs from the collected data.

The field of catalysis research is advancing through the application of machine learning.

By predicting catalytic performance on the basis of experimental data, research time and costs can be reduced.

However, inconsistent data formats and varied terminology complicate the merging of datasets to obtain a sufficiently large dataset for machine learning.

With ontology-driven data capture, we aim to reduce ambiguity in the interpretation of experimental data, thereby enhancing the comparability of datasets for machine learning.

We have found that combining a modular approach to modeling experimental data in an ontology with the reuse of existing ontologies allows for a structured way to capture local experimental setups while ensuring data compatibility beyond the local lab.

This is demonstrated through the example of an ontology for capturing experimental data for industrially relevant selective oligomerization catalysis.

Global change negatively impacts ecosystems, biodiversity and agriculture. Intensive research efforts with numerous case studies in these fields generate large amounts of datasets that should be combined and comprehensively analyzed to synthesize new overarching findings. This requires very good research data management practices according to the FAIR data principles. Interdisciplinary projects should carefully use ontologies for semantic annotation to reduce terminology and data integration problems. Before developing new domain or application specific ontologies, suitable existing ontologies for reuse must be identified. This is a challenging task due to the ever-growing and complex landscape of new ontologies as well as various domain-specific ontology portals that should but may not simplify ontology search. Here, we provide an overview of ontology portals and ontologies, in interdisciplinary ecosystem, biodiversity and agricultural research, and discuss open challenges for ontology search and development.

In biodiversity research, synthesizing data from different sources is frequently needed as a prerequisite to answering important questions. Performing this synthesis and integrating one's own research data remains a tedious process requiring significant human effort. Often, the results of these efforts are not easily reusable for other questions. Knowledge graphs have been proposed in the literature as an approach to alleviate this problem, in part through their inherent adherence to the FAIR data principles, but have gained little traction in biodiversity research practice so far due to significant challenges in knowledge graph construction and usage by non-domain experts. Our contribution showcases an approach and tools needed for knowledge graph creation, management, and usage implemented in the context of PlantHub and the former iKNOW project. We present a knowledge graph combining plant trait sources within the PlantHub project (planthub.idiv.de) including preprocessed data from TRY, a plant trait database, with citizen science occurrence data from naturgucker.de and add taxonomic and additional information from multiple sources (e.g. Wikidata, GBIF, OpenElevation,...). We present the workflow needed to create such a graph and show different options for its management using features from Ontotext Refine for data cleaning, API \& URL fetching, RDF mapping and export, and Ontotext GraphDB for hosting, querying, and visualization. To simplify usage of this graph, we showcase a query builder interface that allows users to construct SPARQL queries without needing any prior domain knowledge. We motivate our work and its application in the biodiversity research domain and contribute to bridging the gap towards using semantic technologies in this field of research.