727 research outputs found
Sort by
ACL word segmentation correction
No full textThe data in this collection consists of two parallel directories, one ("raw") containing the raw text of 18850 articles from the ACL 2013/02 collection, the other ("re-segmented") the word-resegmented version of these articles, obtained using nematus, a seq2seq neural model used for machine translation. The motivation for the work was that spurious spaces in the text seemed to be very common, particularly in older papers, obtained by OCR-ing scanned papers
Unravelling the formation of Hells Bells: underwater speleothems from the Yucatán Peninsula [Data]
No full textHells Bells are unique bell-shaped underwater speleothems recently discovered in deep meromictic sinkholes (cenotes) on the Yucatán Peninsula, Mexico. In order to unravel the underwater growth of Hells Bells and identify the mechanism of subaqueous calcite precipita-tion, several deep meromictic cenotes with and without Hells Bells were investigated. Compre-hensive hydro-, bio- and geochemical analyses of the water bodies and Hells Bells specimens including stable carbon isotope analyses, petrographic examination, U series age-dating, aqui-fer monitoring and microbiological investigations, resulted in a consistent hypothesis on Hells Bells formation. The subaqueous authigenic calcite precipitation in cenotes with Hells Bells is most likely promoted by biogeochemical processes within the narrow 1–2 m thick pelagic re-doxcline immediately above the halocline. Chemolithoautotrophy and incomplete microbial oxidation of sulfide to zero-valent sulfur via proton-consuming sulfide oxidation were identi-fied as dominant biogeochemical processes within the redoxcline. Both processes favor authi-genic calcite precipitation by increasing the pH of the surrounding water. A low degree of con-vection within the water body is identified as crucial hydrogeological prerequisite for the devel-opment of pelagic redoxclines, in which biogeochemical conditions favoring calcite precipita-tion prevail. Furthermore, in order to preserve the authigenic calcite formed in the redoxcline and build Hells Bells, a dynamic elevation of the halocline is hypothesized. Suitable conditions for Hells Bells formation are likely to be encountered in other deep stratified cenotes on the Yucatán Peninsula that are so far unexplored or unrecognized. Age-dating indicates that Hells Bells growth is an ongoing process covering the present to previous interglacial phases from ~0.1 12.5 and ~90–120 ka BP, respectively, outlasting aerial exposure during glacial sea level low stands. Eventually, isotope-geochemical and age-dating analyses reveal a great potential of Hells Bells as a geoarchive for reconstructing minimum sea level elevations, the development of the Yucatán Karst Aquifer and vegetation changes on the Yucatán Peninsula
Do Injunctive or Descriptive Social Norms Elicited Using Coordination Games Better Explain Social Preferences? [Dataset]
No full textWe experimentally study the relationship between social norms and social preferences on the individual level. Subjects coordinate on injunctive and descriptive norms, and we test which type of norm is more strongly related to behavior in a series of dictator games. Our experiment yields three insights. First, both injunctive and descriptive norms explain dictator behavior and recipients´ guesses, but perceptions about descriptive social norms are behaviorally more relevant. Second, our findings corroborate that coordination games are a valid tool to elicit social norm perception on the subject level, as the individuals´ coordination choices are good predictors for their actual behavior. Third, average descriptive norms on the population level accurately predict behavior on the population level. This suggests that the elicitation of descriptive social norms using coordination games is a potentially powerful tool to predict behavior in settings that are otherwise difficult to explore
Distance from a Distance: The Robustness of Psychological Distance Effects [Dataset]
No full textPsychological distance effects have attracted the attention of behavioral economists in the context of descriptive modeling and behavioral policy. Indeed, psychological distance effects have been shown for an increasing number of domains and applications relevant to economic decision making. The current paper questions whether these effects are robust enough for economists to apply them to relevant policy questions. We demonstrate systematic replication failures for the distance-from-a-distance effect shown by Maglio et al. (2013, Journal of Experimental Psychology: General), and relate them to theoretical arguments suggesting that
psychological distance theories are currently too poorly specified to make predictions that are precise enough for economic analyses
HeiCuBeDa Hilprecht - Heidelberg Cuneiform Benchmark Dataset for the Hilprecht Collection
No full textThe number of known cuneiform tablets is assumed to be in the hundreds of thousands. A fraction has been published by printing photographs and manual tracings in books, which is collected by the online Cuneiform Digital Library Initiative (CDLI) catalog including some of these images and providing metadata for more than 100.000 tablets. While 3D-acquisition of tablets is the most modern way for their documentation, the number of 3D-datasets is relatively small and often not openly accessible. However, the Hilprecht Archive Online (HAO) provides 1977 high-resolution 3D scans of tablets under an Open Access license. While both the HAO and the CDLI are accessible publicly, large-scale machine learning and pattern recognition on cuneiform tablets remains elusive, because the data is only accessible by navigating web pages, the tablet identifiers between collections are inconsistent, and the 3D data is unprepared and challenging for automated processing. We enable large-scale analysis of cuneiform tablets by this HeiCuBeda for Hilprecht assembly, which is a cross-referenced benchmark dataset of processed cuneiform tablets: (i) frontally aligned 3D tablets with pre-computed high-dimensional surface features, (ii) six-views raster images for off-the-shelf image processing, and (iii) metadata, transcriptions, and transliterations, for a subset of 707 tablets, for learning alignment between 3D data, image and linguistic expression. This is the first dataset of its kind, and of its size, in cuneiform research. This benchmark dataset is prepared for ease-of-use and immediate availability for computational researches, lowering the barrier to experiment and apply standard methods of analysis. A script in Python is provided to retrieve and compute an updated JSON database of the CDLI’s metadata and raster images. Up-to-date code and meta-data are also available at https://gitlab.com/fcgl/releases/-/tree/master/mara_icdar_2019
Sentiment View Lexicon (EN)
No full textThis gold standard contains sentiment expressions (verbs, nouns and adjectives) that have been annotated according to their (prior) sentiment view. Each sentiment expression is labelled either as actor or speaker view
SRL4ORL: Improving Opinion Role Labeling Using Multi-Task Learning With Semantic Role Labeling [Source Code]
No full textThis repository contains code for reproducing experiments done in Marasovic and Frank (2018).
Paper abstract:
For over a decade, machine learning has been used to extract opinion-holder-target structures from text to answer the question "Who expressed what kind of sentiment towards what?". Recent neural approaches do not outperform the state-of-the-art feature-based models for Opinion Role Labeling (ORL). We suspect this is due to the scarcity of labeled training data and address this issue using different multi-task learning (MTL) techniques with a related task which has substantially more data, i.e. Semantic Role Labeling (SRL). We show that two MTL models improve significantly over the single-task model for labeling of both holders and targets, on the development and the test sets. We found that the vanilla MTL model, which makes predictions using only shared ORL and SRL features, performs the best. With deeper analysis, we determine what works and what might be done to make further improvements for ORL.
Data for ORL
Download MPQA 2.0 corpus.
Check mpqa2-pytools for example usage.
Splits can be found in the datasplit folder.
Data for SRL
The data is provided by: CoNLL-2005 Shared Task, but the original words are from the Penn Treebank dataset, which is not publicly available.
How to train models?
python main.py --adv_coef 0.0 --model fs --exp_setup_id new --n_layers_orl 0 --begin_fold 0 --end_fold 4
python main.py --adv_coef 0.0 --model html --exp_setup_id new --n_layers_orl 1 --n_layers_shared 2 --begin_fold 0 --end_fold 4
python main.py --adv_coef 0.0 --model sp --exp_setup_id new --n_layers_orl 3 --begin_fold 0 --end_fold 4
python main.py --adv_coef 0.1 --model asp --exp_setup_id prior --n_layers_orl 3 --begin_fold 0 --end_fold 10<p
Ambiguity Attitudes in the Loss Domain: Decisions for Self versus Others [Dataset]
No full textWe study whether people’s ambiguity attitudes differ when deciding for themselves or for others in the loss domain. We find no differences in ambiguity attitudes between self- and other-regarding decision-making. Our results are consistent with the loss part of the fourfold pattern of ambiguity attitudes
Lauchheim II.1. Katalog der Gräber 1–300 [Supplement]
No full textMit rund 1300 Gräbern aus dem Zeitraum vom späten 5. bis zum späten 6. Jahrhundert ist das Gräberfeld von Lauchheim 'Wasserfurche' (Ostalbkreis) bis heute der größte bekannte merowingerzeitliche Bestattungsplatz Süddeutschlands. In den Jahren 1986 bis 1996 wurde das fast vollständig erhaltene Gräberfeld großflächig ausgegraben. Seit 2009 widmete sich ein multidisziplinäres Forschungsprojekt der Dokumentation und Auswertung der Lauchheimer Gräber und Grabbeigaben. Dabei kamen in großem Maßstab innovative Dokumentationsmethoden mit dem Ziel einer möglichst vollständigen Datenerfassung zum Einsatz. Als Ergebnis dieses Projekts werden die Befunde und Funde aus Lauchheim in einem detaillierten Katalog mit umfangreichem Tafelteil vorgelegt und damit der weiteren wissenschaftlichen Auswertung zugänglich gemacht. Der vorliegende erste von insgesamt vier Teilbänden des Katalogs umfasst die Gräber 1–300.
Als Datensupplement werden Abbildungen der Perlen aus den Gräbern bereitgestellt. Die Abbildungen liegen als separate tif-Dateien sowie zusammengefasst in einer pdf-Datei vor. Die Dateinamen der einzelnen Bilddateien setzen sich aus der Grabnummer, der Objektnummer und einer Bildnummer zusammen
GECCA mapped
No full textGECCA mapped is a pilot project that visualizes and provides geo-referential metadata of sixty exhibition entries collected in the larger GECCA data base (more than 700 entries). The exhibition sample is limited to Western, i.e. Western European and Northern American group exhibitions, and excludes bi-/ triennials. With the support of the HRA (Heidelberg Research Architecture), GECCA mapped allows the user to trace the exhibition sample implemented in Google Earth. The GECCA mapped logo indicates the place where a particular exhibition was staged and is scaled according to the number of participating artists. A click on the logo opens a pop-up window presenting more information on the exhibition. The Google Earth timeline enables the user to follow the exhibition development in any chosen geographical area in the period from 1982 (earliest exhibition entry) to 2009 (latest exhibition entry).
Group Exhibitions of Contemporary Chinese Art (GECCA): The medium of (group and panoramic) exhibitions has played a fundamental role in creating a global context for Chinese art within and outside of the People's Republic after the end of the "Great Proletarian Cultural Revolution" (1966-1976) and since the political reforms initiated by Deng Xiaoping in 1978/79. In economic, discursive, aesthetic and institutional terms, the Western reception of these shows was very influential for the establishment of a certain international canon of artworks, artists and curators. This particular canon in fact came to be considered representative of the whole of Chinese artistic production, although it actually tends to exclude large parts of the overall artistic activity such as "national ink painting" (guohua), conventional or conservative academic oil painting, as well as those works involving political or consumption oriented subject matter, including mass-produced decorative and popular artworks.
With 60 exhibitions entries, the data that GECCA mapped visualizes is a comparatively small sample of the database GECCA - which contains more than 700 exhibition entries. The data was individually researched and includes information on the location, institution, dates, exhibition topic, participating artists and curators. The sources for the data stem from exhibition catalogues, museum websites, archival documentation of public art libraries and other archives.
A typical use of the kmz-file that visualizes GECCA mapped is Google Earth