The AWA group researches practice-relevant methods and develops modular software systems for the analysis of unstructured data and data streams with a special focus on text messages in social media. Various online platforms, such as Twitter or Reddit, enable active participation in global public discourse. The data generated in this way are often freely accessible and can provide valuable information for a wide range of applications.
The group's application portfolio currently addresses the areas of civil security, natural hazards (natural disasters, extreme weather events, major disasters and the current pandemic) and other environmental impacts (e.g. noise) - but is not limited to these.
Depending on the application and the question at hand, the first step is to identify the right data sources and to evaluate these according to their potentials and risks. The application-relevant data to be extracted from the often very large amounts of data are collected using adapted procurement strategies or filtering and classification methods. Content (e.g. texts and images), relations (e.g. comments, reactions, links to other platforms) and metadata (e.g. time stamp and geolocation) form the basis for a wide range of thematic and spatiotemporal analyzes. Approaches of supervised and unsupervised machine learning, computational linguistics and geostatistics are used.
Current application-oriented research topics: