AI4DiTraRe: Studying Applied AI Within the Process of Digitalisation of Research

The recently established Leibniz Science Campus “Digital Transformation of Research” (DiTraRe) investigates the effects of the broadly understood process of digitalisation of research on a multilevel scale. The project concentrates on four research clusters concerning different topics and gathering use cases from varying scientific areas. For a multiscale investigation these research clusters are interwoven with four dimensions, which approach the tasks from different perspectives and pose their own research questions. Within this matrix we are not only developing practical solutions for each use case but also seeking to find generalisations valuable to the scientific community as well as society in general. To gain access to the abundance of information that is available today, search engines and sophisticated information processing applications are required. In order to obtain well-structured knowledge, technologies such as natural language processing, knowledge extraction, and ontology engineering must be applied. Semantic technologies provide a formal representation of knowledge contained in research data, thus facilitating the efficient integration of heterogeneous data sources. The growing adoption of AI-based knowledge mining technologies requires comprehensible and trustworthy AI algorithms (“explainable AI”). Both statistic and linguistic analysis methods as well as machine learning in combination with symbolic logic and interference mechanisms are applied. The aforementioned AI technologies, among others, are investigated within the DiTraRe project by the dimension “Exploration and knowledge organisation”. For the use case “Sensitive data in sports science”, which is encapsulated in the research cluster “Protected data spaces”, our plan is to develop a knowledge graph which will enable sports scientists to easily analyse their fitness data and make predictions of their studies’ outcomes. We will significantly enhance the ability to interpret fitness data for sports research. In the cluster “Smart data acquisition” we are working with chemists on novel methods of data acquisition. This includes partial automatisation as further digitalisation of a chemistry lab within the Chemotion Electronic Lab Notebook. The use case “Artificial Intelligence in Biomedical Engineering”, enclosed in the cluster “AI-based knowledge realms”, will profit from our support concerning introducing large language models into their research. With biomedical engineers we will use state-of-the-art AI techniques to develop a novel method of predicting the length of stay at an intensive care unit in a non-invasive and much quicker way. We are cooperating with climate researchers on the use case of “Publication of large datasets” (as a representation of “Publication cultures” cluster) where we are employing AI techniques for an unchallenging organisation of very large amounts of research data. A feature to support the process of creation of a uniform platform which we will construct with climate researchers will strongly increase the re-use and availability of earth science data.