• Digital Humanities and Cultural Heritage
• Open Access close
• Research data close
• Dataset close
• Publications at Bielefeld Universit... close

This dataset comprises 250 natural language questions over DBpedia 3.9, annotated with SPARQL queries and answers. It was part of the QALD-4 open challenge: questions 1-200 constitute the training question set and questions 201-250 constitute the test question set.

This data publication consists of two parts. First, it contains a C++ implementation of models of spatial language understanding. More specifically, the source code implements - the Attentional Vector Sum (AVS) model proposed by Regier and Carlson (2001), - the reversed AVS model proposed by Kluth, Burigo, and Knoeferle (2017), - the AVS-BB (AVS bounding box) and the rAVS-CoO (rAVS center-of-object) model proposed by Kluth, Burigo, Schultheis, and Knoeferle (accepted 2018), - and a number of closely related modifications. The source code is fully documented and accompanied by instructions and empirical data to replicate the model simulations reported in Kluth, Burigo, Schultheis, and Knoeferle (accepted 2018) and Kluth and Schultheis (2018). Second, this data publication contains the empirical data (rating data, eye movement data, and reaction time data) from the study described in Kluth, Burigo, Schultheis, and Knoeferle (accepted 2018). Furthermore, R source code files with statistical analyses (mainly Bayesian regression models) are provided that can be used to replicate all statistical models and graphics reported in Kluth, Burigo, Schultheis, and Knoeferle (accepted 2018). All source code is licensed under the GPLv3. All data are licensed under the ODbL v1.0.

This corpus consists of annotations of Amazon reviews for different product categories in the languages German and English. The reviews themselves are not part of this data publication. The annotations are fine-grained, including aspects and subjective phrases. In addition, the relation of an aspect to be a target of a subjective phrase is provided as well as the polarity of the subjective phrase. The corpus consists of 622 English and 611 German reviews for coffee machines, cuterly, microwaves, toaster, trash cans, vacuum cleaner, washing machines and dishwasher. The English corpus is annotated with more than 8000 aspects and 5000 subjective phrases, the German part with more than 6000 aspects and around 5000 subjective phrases (depending on the annotator). Each review is independently annotated by two annotators. Updates to these data will be available at [http://www.roman-klinger.de/usagecorpus](http://www.roman-klinger.de/usagecorpus).

This dataset comprises all questions used as benchmark in the 5th Open Challenge on Question Answering over Linked Data (QALD-5). Questions 1-340 and 391-410 are the training questions for multilingual question answering over DBpedia and hybrid question answering, respectively, and questions 341-390 and 411-420 are the corresponding test questions. [Documentation]: https://github.com/ag-sc/QALD/blob/master/5/documents/qald-5.pdf [Documentation]:

This dataset comprises 200 natural language questions over DBpedia 3.8, annotated with SPARQL queries and answers. It was part of the QALD-3 open challenge: questions 1-100 constitute the training question set and questions 101-200 constitute the test question set.