piwik-script

Deutsch Intern
    Data Science Chair

    Projects

    We apply our data science, machine learning und deep learning research to several different application areas, which are shown below. Each application area lists the related running and concluded projects on a seperate page.

    Natural Language Processing

    Our research in the area of Natural Language Processing ranges from scene detection in literature to sentiment analysis in chat data.

    Ecosystems and Climate Modeling

    We use machine learning and data science methods to provide new modeling approaches or substitute existing statistical models, focusing on climate models, environmental factors and bee behaviour.

    Publication Data

    In several projects we have researched semantics and connections between publications and with BibSonomy,  we released a bookmarking platform for publications. 

     

     

    Product and Chat Recommendation

    We are working on recommender systems for various applications like e-commerce shops, but also for chat interactions,  clinical diagnostics or publication data.

    Physics Informed Deep Learning

    In several projects, we aim to apply deep learning methods to model physical systems. We use these to model electro magnetic fields for magnetic resonance imaging or to model the structural integrity of bridges.

    Network Security & Fraud Detection

    We apply machine learning-based approaches for anomaly detection and develop novel ML methods specifically tailored to the domains of network security and fraud detection.

    Medical & Biological Data

    We apply machine learning techniques to work with medical data from diagnostics and biological data for single cell analysis.