idw – Informationsdienst Wissenschaft
Nachrichten, Termine, Experten
Nachrichten, Termine, Experten
idw - Informationsdienst
Wissenschaft
Chemists at LIKAT in Rostock have developed a digital platform that links data from catalysis research and makes it available to experts for research and exchange, as well as for future AI applications in this complex discipline. At the heart of the cloud-based database is the digital dictionary Voc4Cat. It defines relevant terms in a user-friendly and machine-readable way, for example on procedures, starting materials, and products of laboratory work, as well as on analytics and modeling. Voc4Cat is available to the community as open source via GitHub.
As with Wikipedia, registered users can contribute to this digital knowledge repository by gradually feeding it with data. The system was developed at the Leibniz Institute for Catalysis in Rostock with DFG funding as part of the NFDI4Cat consortium, which is part of the National Research Data Infrastructure (NFDI). It was developed over the past two years by a team led by LIKAT chemist Dr. David Linke. David Linke also programmed the first vocabulary for Voc4Cat, which currently comprises around 500 terms, together with Dr. Nikolaos Moustakas.
Catalysis as a key technology
Catalysis is currently developing into a key technology, for example for the development of a sustainable and, above all, climate-friendly energy supply. This is reflected in the rapidly growing number of publications and a flood of data. Dr. Moustakas: "Until now, each research team has used its own terms when documenting and publishing research data. This complicates the increasingly AI-supported linking of research data."
According to him, it was high time to create a basis for networking research results in catalytic chemistry and related fields such as chemical engineering and process engineering. The open-source project enables the community to jointly maintain and expand the vocabulary by providing access to a central digital repository.
FAIR criteria for data management
The FAIR criteria – findability, accessibility, interoperability, and reusability – are the global standard for the digital management of research data. Interoperability means preparing and documenting research data in such a way that machines can read it clearly and exchange and link it with each other. This is exactly what Voc4Cat contributes to.
"Data compatibility is a particular challenge for programmers, because even in the so-called exact sciences, technical terms are often used differently," says Nikolaos Moustakas. That is why, in addition to the definition, each term in Voc4Cat has an internationally unique identification number, or IRI for short. Information about authors and other sources is also stored, as well as references to synonyms and content-related connections to other methods and concepts.
Compatible with Local Networks and ChemCatChem
Anyone who wants to publish the results of their research on chemical catalysis can now link key terms in their paper to Voc4Cat and assign a defined meaning to "their" formulation. Such networking makes it easier for authors to be found online with their new findings and to be linked to other important topics, which in turn influences the much-vaunted citation rate of a paper.
In collaboration with the journal ChemCatChem from Wiley Publishing, the most important keywords have already been included in Voc4Cat. All information and vocabulary in Voc4Cat is standards-compliant and machine-readable, making it easy to integrate into other software systems, electronic lab journals, or research data portals.
Users can find instructions for making their own entries in the Catalysis Wiki or for suggesting changes to the existing vocabulary at the NFDI4Cat Consortium's GitHub link. Voc4Cat curators can provide support if necessary, and in any case they check the entries for technical accuracy. "Everything is open, anyone can follow what we are doing," says Nikolaos Moustakas, who, like David Linke, is one of the curators.
AI Makes Papers Voc4Cat-compatible
Five years ago, the National Research Data Infrastructure, or NFDI e.V. for short, was founded by the federal and state governments to promote the digitization of data handling. Dr. Moustakas is currently training his own language AI solution, which chemists can use to prepare their already published papers for the digital dictionary. The tool will then recognize which terms it needs to access in these texts – a Voc4Cat-GPT, so to speak. It is expected to be available at the end of the year.
Dr. David Linke, Head of Research Department “Catalyst Discovery & Technical Engineering”: David.Linke@catalysis.de,
Dr. Nikolaos Moustakas, Scientist: Nikolaos.Moustakas@catalysis.de
Link to Voc4Cat: https://nfdi4cat.github.io/voc4cat/
Voc4Cat was developed under his leadership: Dr. David Linke
Quelle: LIKAT
Copyright: LIKAT
Developing a proprietary language AI solution for Voc4Cat: Dr. Nikolaos Moustakas from LIKAT
Quelle: private
Copyright: LIKAT
Voc4Cat logo
Merkmale dieser Pressemitteilung:
Journalisten, Studierende, Wissenschaftler
Chemie, Gesellschaft, Informationstechnik
überregional
Forschungsergebnisse, Wissenschaftliche Publikationen
Englisch
Sie können Suchbegriffe mit und, oder und / oder nicht verknüpfen, z. B. Philo nicht logie.
Verknüpfungen können Sie mit Klammern voneinander trennen, z. B. (Philo nicht logie) oder (Psycho und logie).
Zusammenhängende Worte werden als Wortgruppe gesucht, wenn Sie sie in Anführungsstriche setzen, z. B. „Bundesrepublik Deutschland“.
Die Erweiterte Suche können Sie auch nutzen, ohne Suchbegriffe einzugeben. Sie orientiert sich dann an den Kriterien, die Sie ausgewählt haben (z. B. nach dem Land oder dem Sachgebiet).
Haben Sie in einer Kategorie kein Kriterium ausgewählt, wird die gesamte Kategorie durchsucht (z.B. alle Sachgebiete oder alle Länder).
