University of Mannheim

About: University of Mannheim is a education organization based out in Mannheim, Germany. It is known for research contribution in the topics: Context (language use) & Politics. The organization has 4448 authors who have published 12918 publications receiving 446557 citations. The organization is also known as: Uni Mannheim & UMA.

Technology Enhanced Formative Assessment for 21st Century Learning

Abstract: This article is based on the deliberations of the Assessment Working Group at EDUsummIT 2015 in Bangkok, Thailand. All of the members of Thematic Working Group 5 (TWG5) have contributed to this synthesis of potentials, concerns and issues with regard to the role of technology in assessment as, for and of learning in the 21st century.

Market Structure, Informational Efficiency and Liquidity: An Experimental Comparison of Auction and Dealer Markets

Abstract: This paper reports the results of 18 market experiments that were conducted in order to compare the call market, the continuous auction and the dealer market. The design incorporates asymmetric information but guarantees that the ex-ante quality of the private signals of all traders is identical. Therefore, the aggregation of diverse information can be analyzed in the absence of insider trading. Single transaction prices in the call and continuous auction market are found to be much more efficient than prices in the dealer market. The latter is, however, very efficient when average prices are analyzed. Averaging the prices of a trading period largely eliminates the bid-ask spread. The conclusion is therefore that prices in a dealer market convey high quality information, but at the expense of high transaction costs. The call market, although exhibiting small pricing errors, shows a systematic tendency towards underadjustment to new information. An analysis of market liquidity using various measures proposed in the literature shows that execution costs are lowest in the call market and highest in the dealer market. The analysis also reveals that both the trading volume and Roll's (1984) serial covariance estimator are inappropriate measures of execution costs in the present context. The quality of the private signals traders receive influences portfolio structure but does not influence end-of-period wealth. This result is consistent with efficient price discovery in the experimental markets.

Central bank communication on financial stability

Abstract: Central banks regularly communicate about financial stability issues. This article asks how such communications affect financial markets, based on a unique dataset covering more than 1,000 releases of Financial Stability Reports (FSRs) and speeches by 37 central banks over the past 14 years. The findings suggest that optimistic FSRs lead to significant and potentially long-lasting positive abnormal stock market returns, whereas no such effect is found for pessimistic FSRs. Speeches and interviews, in contrast, have smaller effects on market returns during tranquil times but have been influential during the 2007–10 global financial crisis.

Automatic Cardiac Disease Assessment on cine-MRI via Time-Series Segmentation and Domain Specific Features

Abstract: Cardiac magnetic resonance imaging improves on diagnosis of cardiovascular diseases by providing images at high spatiotemporal resolution. Manual evaluation of these time-series, however, is expensive and prone to biased and non-reproducible outcomes. In this paper, we present a method that addresses named limitations by integrating segmentation and disease classification into a fully automatic processing pipeline. We use an ensemble of UNet inspired architectures for segmentation of cardiac structures such as the left and right ventricular cavity (LVC, RVC) and the left ventricular myocardium (LVM) on each time instance of the cardiac cycle. For the classification task, information is extracted from the segmented time-series in form of comprehensive features handcrafted to reflect diagnostic clinical procedures. Based on these features we train an ensemble of heavily regularized multilayer perceptrons (MLP) and a random forest classifier to predict the pathologic target class. We evaluated our method on the ACDC dataset (4 pathology groups, 1 healthy group) and achieve dice scores of 0.945 (LVC), 0.908 (RVC) and 0.905 (LVM) in a cross-validation over the training set (100 cases) and 0.950 (LVC), 0.923 (RVC) and 0.911 (LVM) on the test set (50 cases). We report a classification accuracy of 94% on a training set cross-validation and 92% on the test set. Our results underpin the potential of machine learning methods for accurate, fast and reproducible segmentation and computer-assisted diagnosis (CAD).

Methods and rules

Abstract: We show how classical datalog semantics can be used directly and very simply to provide semantics to a syntactic extension of datalog with methods, classes, inheritance, overloading and late binding. Several approaches to resolution are considered, implemented in the model, and formally compared. They range from resolution in C++ style to original kinds of resolution suggested by the declarative nature of the language. We show connections to view specification and a further extension allowing runtime derivation of the class hierarchy.