Logo des Repositoriums

Auflistung nach Autor:in "Rehm, Georg"

1 - 4 von 4
  • Konferenzbeitrag
    NFDI4DS Infrastructure and Services
    (INFORMATIK 2023 - Designing Futures: Zukünfte gestalten, 2023) Schimmler, Sonja; Wentzel, Bianca; Bleier, Arnim; Dietze, Stefan; Karmakar, Saurav; Mutschke, Peter; Kraft, Angelie; Taffa, Tilahun A.; Usbeck, Ricardo; Boukhers, Zeyd; Auer, Sören; Castro, Leyla J.; Ackermann, Marcel R.; Neumuth, Thomas; Schneider, Daniel; Abedjan, Ziawasch; Latif, Atif; Limani, Fidan; Abu Ahmad, Raia; Rehm, Georg; Attar Khorasani, Sima; Lieber, Matthias
    NFDI4DataScience (NFDI4DS) is a consortium founded to support researchers in all stages of the research data lifecycle in order to conduct their research in line with the FAIR principles. The infrastructure developed targets researchers from a wide range of disciplines working in the field of data science and artificial intelligence. NFDI4DS contributes to systematically understanding the needs and challenges of researchers in various disciplines regarding data science and artificial intelligence, keeping in mind ethical, legal and social aspects. The identified needs will be addressed by support structures such as educational videos and challenges. Transparency, reproducibility and FAIRness will be improved by integrating existing and newly developed services into the NFDI4DS infrastructure, and by systematically adding all digital objects (articles, data, machine learning models, workflows, scripts/code, etc.) to the NFDI4DS research knowledge graph. This paper presents the goals of NFDI4DS, and gives an overview on what the consortium is going to contribute to the data science and artificial intelligence communities. It focuses on existing and newly developed services and their integration.
  • Konferenzbeitrag
    NFDI4DS Shared Tasks
    (INFORMATIK 2023 - Designing Futures: Zukünfte gestalten, 2023) Abu Ahmad, Raia; Borisova, Ekaterina; Rehm, Georg; Dietze, Stefan; Kamakar, Saurav; Otto, Wolfgang; D’Souza, Jennifer; Limani, Fidan; Usbeck, Ricardo
    Shared tasks have proven to be successful in proposing innovative solutions for challenging research problems. The NFDI4DS consortium plans to host various shared tasks to tackle problems under the umbrella of scholarly information processing. We discuss three shared tasks in detail: Field of Research Classification, Software Mention Detection, and Tracking State-of-the-Art in Empirical AI. We also briefly mention other shared tasks planned to be released in the future.
  • Konferenzbeitrag
    NFDI4DS Transfer and Application
    (INFORMATIK 2023 - Designing Futures: Zukünfte gestalten, 2023) Borisova, Ekaterina; Abu Ahmad, Raia; Rehm, Georg; Usbeck, Ricardo; D’Souza, Jennifer; Stocker, Markus; Auer, Sören; Gilsbach, Judith; Wolschewski, Anastasia; Keller, Johannes; Schneider, Daniel; Neumuth, Thomas; Schimmler, Sonja
    Due to the ever increasing importance of Data Science and Artificial Intelligence methods for a wide range of scientific disciplines, ensuring transparency and reproducibility of DS and AI methods and research findings have become essential. The NFDI4DS project promotes the findability, accessibility, interoperability, and reusability in DS and AI by developing an open integrated research data infrastructure in which all artefacts (e. g., papers, code, models, datasets) will be interlinked in a FAIR and transparent way. One of the key aspects is to build a bridge between NFDI4DS and other research communities which actively apply DS and AI methods. This paper describes the main actions taken to engage with the relevant (sub)communities.
  • Konferenzbeitrag
    Research Knowledge Graphs in NFDI4DS
    (INFORMATIK 2023 - Designing Futures: Zukünfte gestalten, 2023) Karmakar, Saurav; Zloch, Matthäus; Limani, Fidan; Zapilko, Benjamin; Upadhyaya, Sharmila; D’Souza, Jennifer; Castro, Leyla J.; Rehm, Georg; Ackermann, Marcel R.; Sack, Harald; Boukhers, Zeyd; Schimmler, Sonja; Dessí, Danilo; Mutschke, Peter; Dietze, Stefan
    The ever-increasing amount of research output through scientific articles requires means to enable transparency and a better understanding of key entities of the research lifecycle, referred to as research artifacts, such as methods, software, datasets, etc. Research Knowledge Graphs (RKG) make research artifacts findable, accessible, interoperable, and reusable (FAIR) and facilitate their interpretability. In this article, we describe the role of RKGs, from their construction to the expected benefits, including an overview and a vision of their use within the German National Research Data Infrastructure (NFDI) consortium NFDI4DataScience (NFDI4DS). This paper includes insights about the existing RKGs, how to formally represent research artifacts, and how this supports better transparency and reproducibility in data science and artificial intelligence. We also discuss key challenges, such as RKG construction, and integration, and give an outlook on future work.