Navigation auf


Department of Informatics Interactive Visual Data Analysis Group

Interactive Visual Data Analysis Group

Welcome to the Interactive Visual Data Analysis Group

Image Gallery

Our interdisciplinary research focus is at the intersection of Information Visualization, Visual Analytics, HCI, and Machine Learning, emphasizing a human-centered approach to data science and explainable AI. We address complex challenges across data, model, and human interaction, employing techniques from unsupervised to supervised learning in a human-in-the-loop approach. Many of our contributions include aspects of data exploration, model explanation, and human knowledge externalization. Our work aims to enhance data-driven insight generation and decision-making in fields like digital health, humanities, and libraries by facilitating user engagement with complex data through interactive, visual, and personalized machine learning environments.


|News | Prof. Jürgen Bernard | Team | Research | Teaching | Open Positions |

Challenge Areas

Data-oriented challenges arise from different types of complexity data comes with. Examples are heterogeneous data, dirty data, uncertain data, or unlabeled data. Many of our solution provide users with sophisticated data exploration capability, to master data-driven decision-making.

Model-oriented challenges refer to the need for an effective and efficient use of algorithmic models to cope with data challenges. Often, several models are included in a data analysis process. Example challenges include data preprocessing, model building, model quality assessment, or model explanation. Many of our solutions contribute to enhanced model explainability and transparency.

Human-oriented challenges refer to human aspects in data analysis and are particularly interesting for the research focus of IVDA. User-oriented challenges are different degrees of user expertise, users’ personalization intents, understanding and supporting user preferences with respect to data and tasks, as well as human factors in a broader sense. Our design-oriented contributions are tailored to the information need of individual users, with novel approaches that make machine learning and AI applicable to larger user groups in the context of data science and digitalization initiatives.

Weiterführende Informationen