PropBank has been widely used as training data for Semantic Role Labeling. However, because this training data is taken from theWSJ, the resulting machine learning models tend to overfit on idiosyncrasies of that text’s style, and do not port well to other genres. In addition, since PropBank was designed on a verb-by-verb basis, the argument labels Arg2 Arg5 get used for very diverse argument r...

Previous research suggests that three-year-olds fail to learn from statistical data when their prior beliefs conflict with evidence. Are children’s beliefs entrenched in their folk theories, or can preschoolers rationally update their beliefs? Motivated by a Bayesian account, we conducted a training study to investigate this question. Children (45 months) who failed to endorse a statistically m...

Christoph Merkle, Philipp Schreiber and Martin Weber⇑ Merkle is an associate professor of finance at Aarhus University a research fellow the Danish Finance Institute. full Esslingen University. Weber senior Mannheim CEPR. corresponding author: cmerkle{at}econ.au.dk

Many researchers are reluctant to use aggregate data in program evaluation and other policy relevant research. Aggregate data refers to data on individuals that have been averaged by year, by geographic area, by service agency, or in some other way. For example, data on the individual earnings of job training participants may be aggregated by year or student test scores may be aggregated by sch...

The ability of pervasive context-aware systems to perform efficiently relies on their ability to gather full and unambiguous information about the environment. But raw data collected from sensors is often noisy, imprecise and corrupted, which leads to inconsistencies and conflicts in gathered data. Also environment is only partially observable, thus allowing ambiguities in the knowledge about i...