Geo-Social Analytics Lab

The Geo-Social Analytics Lab was a research group at Z_GIS that existed from 2013 to 2024 and was headed by Prof Bernd Resch. He now leads the Geosocial Artificial Intelligence research group at IT:U.

The Geo-social Analytics Lab was concerned with the development of sound scientific methods to derive value in the growing availability of open data, sensor data, transaction or social media data. In particular, the research investigated how behaviour in the real world is reflected in virtual worlds and vice versa. The interdisciplinary research explored a range of technologies which are able to sense, directly or indirectly, a variety of environmental, human and social phenomena. Such sensing technologies generate vast and rapidly increasing volumes of digital sensor data. Several empirical studies were carried out in the research area, with a particular focus on urban areas.

An overview of the publications can be found here.

Projects

AIFER (Artificial Intelligence for Emergency Response) was funded by the FFG’S KIRAS programme, as well as by the German Ministry for Education and Research (BMBF). It was jointly coordinated with the German Aerospace Center (DLR). The aim of the project was to develop artificial intelligence methods to derive better information for disaster management from remote sensing and social media data. A highlight of the project was a large-scale field exercise in April 2023. The project was a collaboration with DLR, Johanniter, the Austrian Red Cross, the Bavarian Red Cross, the German Federal Agency for Technical Relief (THW), IFES, Spatial Services, and Disy.
DEGENET investigated anti-democratic dynamics in geo-social media, including fake news, hate speech, and conspiracy narratives. It aimed to develop analytical methods to identify democracy-threatening content and provide insights for political, educational, and civil society responses. The project laid the groundwork for future regulatory measures and strategies to counteract these developments.
DigitalInnovationLayer (2023-2026) is funded by FFG’s BRIDGE programme and deals with the analysis of innovation through digital data sources. In particular, corporate websites and social media data. ISTARI.AI, a German start-up, is responsible for acquiring and processing data from company websites. Spatial Services, a spin-off of the University of Salzburg, provides expertise in the field of remote sensing, which is an important data source in the project.
E2mC explored the integration of social media analysis and crowdsourced information into Copernicus Emergency Management Service (EMS). It developed a prototype service, Copernicus Witness, to enhance crisis management by leveraging heterogeneous social media data and federated crowdsourcing communities. The project aimed to improve disaster response by reducing temporal delays and increasing situational awareness.
ESSEM (2022 – 2025) was funded by the German Federal Ministry for Digital and Transport (BMDV) and investigated the feasibility of using physiological reactions of the human body, recorded through non-invasive wearable sensing technology, to guide (e)-bike centric planning approaches. The project was a collaboration with KIT, IAT, the city of Osnabrück, Ludwigsburg, and Outdooractive.
GeoSHARING dealt with analysing geo-social media using geospatial machine learning to support humanitarian decision-making. It investigated new cross-disciplinary machine learning methods, dealing with social media data’s noisy characteristics.
HUMAN+ was concerned with the analysis of migration movements from different data sources. In this, the detection and prediction of migration movements, as well as identification of areas that are frequently passed during migrations, was central. The results of HUMAN+ helped to mitigate the problems that arise in the wake of large migration movements, and thus madesuch events more manageable for all parties involved. The project was a collaboration with Joanneum Research, IFES, IFR, Johanniter, Spatial Services, University Graz, Bundesheer, Fraunhofer IAIS, University Kassel, the German Federal Agency for Technical Relief (THW), VOMATEC, CrisCom Solutions, the University Passau, Deutsche Hochschule der Polizei and the Bavarian Red Cross.
Malawi Atlas was a web portal that visualizes geospatial data in the context of disaster management, containing various layers such as UAV images, demographic data, lancover and natural hazards.
Music Map visualises the spatio-temporal distribution of musical practices in the city and province of Salzburg. It was a collaboration with Mozarteum.
MUSIG (Multi-sensor Information Generation to Support Crisis Management and Prevention) was a research project funded by the FFG’s KIRAS programme. It focused on the automated extraction of collective movement information from geo-social media, mobile phone data and in-situ imagery with AI methods, and the scenario-oriented fusion of movement information in a new mixed-methods approach, as well as its provision for crisis management and prevention in near real time including nowcasting information. It was a collaboration with Johanniter, the Austrian Red Cross, Joanneum Research, Spatial Services, and Eurofunk.
Perigee explored social media-based donor identification for emergency fundraising. It developed a prototype platform using machine learning to filter and analyze social media activity, enabling targeted engagement with potential donors through visualization and analytical tools.
Spatio-Temporal Epidemiology of Emerging Viruses (2022-2025): This research project leverages geo-social media, web searches, and official health data to transform the detection and prediction of viral outbreaks. By developing advanced spatio-temporal and AI-driven models, it enhances real-time disease tracking, enabling faster and more accurate responses to threats like dengue, Zika, Chikungunya, and COVID-19. Pioneering novel digital data approaches, this project paves the way for the future of epidemiological analysis and outbreak prevention. It is a collaboration with Heidelberg University Hospital and the Department of Geography of Heidelberg University. The project was transferred to IT:U.
TEMA (2021-2026) stands for „Trusted Extremely Precise Mapping and Prediction for Emergency Management”. It is a four year long project funded by the European Union through its Horizon Europe. The main objectives are improving disaster management by using new digital technologies, accelerating extreme data analytics and increasing its trustworthiness, and improving the prediction and visualisation of disaster-related outputs. The project involves 20 partners from Austria, Belgium, Finland, Germany, Greece, Italy, the Netherlands, and Spain. It is coordinated by the Aristotle University of Thessaloniki. The Geo-Social Analytics Lab contributes a social media component to the proposed systems that analyses disaster-related, georeferenced posts. For more information, see the website ( https://tema-project.eu/) or the project LinkedIn page. The project was transferred to IT:U.
Urban Emotions explored the role of real-time human sensory data and crowdsourced social media insights in urban planning. By analyzing emotions from smartband sensors and geosocial media data, it developed innovative methodologies for understanding human-city interactions. The project provided new visualization techniques and practical recommendations, demonstrated through case studies in Heidelberg, Kaiserslautern, and Boston.
Urban Spaces investigated the origins and properties of intrinsic spatial scales in cities. It analyzed one-dimensional scales through distance decay functions and two-dimensional scales by examining the agglomeration and fragmentation of neighborhoods. The project developed high-performance software for large-scale comparative analyses of major cities, providing a detailed theoretical and empirical understanding of urban spatial structures.
Walk & Feel studied walkability by collecting physiological data from 60 participants in three urban areas. Using smartbands and a smartphone e-diary, it linked biometric responses with location data to analyze how urban environments influence human well-being. The project aimed to enhance existing walkability assessment methods and provide guidelines for their application.