Julian Bruns

BigGIS: a continuous refinement approach to master heterogeneity and uncertainty in spatio-temporal big data (vision paper)

Geographic information systems (GIS) are important for decision support based on spatial data. Due to technical and economical progress an ever increasing number of data sources are available leading to a rapidly growing fast and unreliable amount of data that can be beneficial (1) in the approximation of multivariate and causal predictions of future values as well as (2) in robust and proactive decision-making processes. However, todays GIS are not designed for such big data demands and require new methodologies to effectively model uncertainty and generate meaningful knowledge. As a consequence, we introduce BigGIS, a predictive and prescriptive spatio-temporal analytics platform, that symbiotically combines big data analytics, semantic web technologies and visual analytics methodologies. We present a novel continuous refinement model and show future challenges as an intermediate result of a collaborative research project into big data methodologies for spatio-temporal analysis and design for a big data enabled GIS.

Automated Quality Assessment of (Citizen) Weather Stations

Today we have access to a vast amount of weather, air quality, noise or radioactivity data collected by individual around the globe. This volunteered geographic information often contains data of uncertain and of heterogeneous quality, in particular when compared to official in-situ measurements. This limits their application, as rigorous, work-intensive data cleaning has to be performed, which reduces the amount of data and cannot be performed in real-time. In this paper, we propose dynamically learning the quality of individual sensors by optimizing a weighted Gaussian process regression using a genetic algorithm. We chose weather stations as our use case as these are the most common VGI measurements. The evaluation is done for the south-west of Germany in August 2016 with temperature data from the Wunderground network and the Deutsche Wetter Dienst (DWD), in total 1561 stations. Using a 10-fold cross-validation scheme based on the DWD ground truth, we can show significant improvements of the predicted sensor reading. In our experiment we were obtain a 12.5% improvement on the mean absolute error.