Search

Police like other public organizations increasingly use social media for external communication. Due to their bureaucratic organization social media poses a communication challenge for them. This study analyses the content of tweets by German police using a three-category framework. Machine-learning based classification tasks are coupled with multilevel modelling to analyse all tweets distributed in 2019. The study demonstrates that police largely use Twitter to distribute information unidirectionally focusing on core tasks, while information gathering, and public relations only play a subordinate role. There is variation between accounts. However, the model only explains parts of it, thus inviting for follow-up research. [ABSTRACT FROM AUTHOR]

The shortage of grafts for liver transplantation requires risk stratification and adequate allocation rules. This study aims to improve the model of end-stage liver disease (MELD) score for 90-day mortality prediction with the help of different machine-learning algorithms. We retrospectively analyzed the clinical and laboratory data of 654 patients who were recruited during the evaluation process for liver transplantation at University Hospital Leipzig. After comparing 13 different machine-learning algorithms in a nested cross-validation setting and selecting the best performing one, we built a new model to predict 90-day mortality in patients with end-stage liver disease. Penalized regression algorithms yielded the highest prediction performance in our machine-learning algorithm benchmark. In favor of a simpler model, we chose the least absolute shrinkage and selection operator (lasso) regression. Beside the classical MELD international normalized ratio (INR) and bilirubin, the lasso regression selected cystatin C over creatinine, as well as IL-6, total protein, and cholinesterase. The new model offers improved discrimination and calibration over MELD and MELD with sodium (MELD-Na), MELD 3.0, or the MELD-Plus7 risk score. We provide a new machine-learning-based model of end-stage liver disease that incorporates synthesis and inflammatory markers and may improve the classical MELD score for 90-day survival prediction. [ABSTRACT FROM AUTHOR]

This study gives an overview of soil organic carbon (SOC) stocks in Germany's agricultural soils, and quantifies and explains the influence of explanatory variables such as land use and management, soil type and climate. Over 2500 agricultural sites were sampled and their SOC stocks determined, together with other soil properties. Machine‐learning algorithms were used to identify the most important variables. Land use, land‐use history, clay content and electrical conductivity were the main predictors in the topsoil, whereas bedrock material, relief and electrical conductivity governed the variation in subsoil carbon stocks. We found that 32% of all soil profiles were anthropogenically transformed. The influence of climate variables was surprisingly small, whereas site variables, in particular in the subsoil, explained a large proportion of the variation in soil carbon. The understanding of SOC dynamics at the regional scale requires a thorough description of the spatial variation in soil properties. The effect of agronomic management on SOC stocks was important near the soil surface, but was mainly attributable to land use (grassland or arable and not to other management factors. Highlights: Factors affecting spatial variation of soil organic carbon stocks are largely unknown.More than 200 possible predictors for this variation were assessed for over 2500 sites.Land use (history), texture, electrical conductivity, bedrock and relief were main controlling factors of carbon stock variation.Aggregation and carbon storage capacity of soil determine much of the spatial variation in carbon stocks. [ABSTRACT FROM AUTHOR]

The fusion of microwave and optical data sets is expected to provide great potential for the derivation of forest cover around the globe. As Sentinel-1 and Sentinel-2 are now both operating in twin mode, they can provide an unprecedented data source to build dense spatial and temporal high-resolution time series across a variety of wavelengths. This study investigates (i) the ability of the individual sensors and (ii) their joint potential to delineate forest cover for study sites in two highly varied landscapes located in Germany (temperate dense mixed forests) and South Africa (open savanna woody vegetation and forest plantations). We used multi-temporal Sentinel-1 and single time steps of Sentinel-2 data in combination to derive accurate forest/non-forest (FNF) information via machine-learning classifiers. The forest classification accuracies were 90.9% and 93.2% for South Africa and Thuringia, respectively, estimated while using autocorrelation corrected spatial cross-validation (CV) for the fused data set. Sentinel-1 only classifications provided the lowest overall accuracy of 87.5%, while Sentinel-2 based classifications led to higher accuracies of 91.9%. Sentinel-2 short-wave infrared (SWIR) channels, biophysical parameters (Leaf Area Index (LAI), and Fraction of Absorbed Photosynthetically Active Radiation (FAPAR)) and the lower spectrum of the Sentinel-1 synthetic aperture radar (SAR) time series were found to be most distinctive in the detection of forest cover. In contrast to homogenous forests sites, Sentinel-1 time series information improved forest cover predictions in open savanna-like environments with heterogeneous regional features. The presented approach proved to be robust and it displayed the benefit of fusing optical and SAR data at high spatial resolution. [ABSTRACT FROM AUTHOR]