This paper presents a platform that uses open urban data and machine learning to predict parking space occupancy in Mainz, Germany. Our goal is to support urban mobility by delivering real-time weather data and parking availability forecasts. We developed and evaluated several machine learning models based on their predictive accuracy. To complement the backend, we designed a visual analytics prototype that supports decision-making. The system provides a user-friendly interface that helps citizens locate available parking more efficiently and reduces traffic caused by parking searches. A user study demonstrates that the platform effectively integrates data-driven forecasts with intuitive visualizations of uncertainty, enhancing user understanding and trust. We designed the prototype to be scalable and adaptable for broader applications in intelligent urban infrastructure.

Uncertainties are frequently encountered in machine learning, especially in classification tasks where the model's output is determined by the probabilities of each class. In geospatial use cases, the challenge is to find appropriate visualizations on geographic maps for these uncertainties, so that domain experts know when to trust the model. This study utilizes the uncertainties of an extreme gradient boosting machine to classify traffic accident severity and presents various visualization profiles. These visualizations are the outcome of an experimental approach. Each profile has its own advantages but may also encounter issues such as overlapping data points or the determination of what uncertain means. The most frequently used visualization types are symbols and colors.

GeoVisual Analytics often uses a wide range of visualization techniques in combination with human interactions. However, in order to improve urban planning decisions and expand the knowledge base about spatio-temporal correlations in cities, we identify high and sometimes difficult-to-overcome data and visualization complexity. Therefore, users need targeted support to reduce this complexity and understand complex multidimensional data. This paper proposes an interactive 3D visualization approach for analyzing spatio-temporal correlations. Effective visual explanations in user-oriented guidance with storytelling elements support this technique. We adapt the approach of flow maps to visualize both resulting correlations and the underlying density of possible destinations from an existing geoanalytical process. Our interactive step-by-step guidance with four narratives aims to create a balance between the author and the user. We evaluated our prototype with various domain experts. The evaluation confirms our approach and improves users' understanding of data and spatio-temporal correlations in urban areas.

This paper deals with the geographical analysis and visualization of network marketing. The aim of the study was to develop interactive visual methods, which help to answer questions related to the analysis of network marketing structures. Those questions were the basis for the research and development performed. The challenges tackled in the paper result from data analysis, which includes a combination of structural data and their geographical position. An approach utilizing interactive drill down was developed. The resulting prototype summarizes the findings and allows the validation of the results.