• Digital Humanities and Cultural Heritage
• Research dataclose
• Research softwareclose
• Other research productsclose
• Open Access close
• Dataset close
• European Commission close
• FI close
• English close

This release contains data sets for experiments with document-level machine translation. The data sets have been used in previous studies and provided here for replicability and comparison with other systems. The data sets are taken from the English-German news translation task at WMT 2019 and the English-German bitext in the OpenSubtitles collection v2016 from OPUS. All data sets are sentence aligned with corresponding lines being aligned to each other. Document boundaries are marked with empty lines (on both sides of the parallel corpus). The data set has been used in the following publication: @inproceedings{scherrer-tiedemann-loaiciga-2019, title = "Analysing concatenation approaches to document-level NMT in two different domains", author = {Scherrer, Yves and Tiedemann, J{\"o}rg and Lo{\'a}iciga, Sharid}, booktitle = "Proceedings of the Third Workshop on Discourse in Machine Translation", month = nov, year = "2019", address = "Hong-Kong", publisher = "Association for Computational Linguistics", } Please, cite that paper if you use the data set in your own work. {"references": ["Scherrer, Tiedemann and Lo\u00e1iciga: \"Analysing concatenation approaches to document-level NMT in two different domains\", in Proceedings of DiscoMT2019 at EMNLP 2019, Hong-Kong"]}

This release contains data sets for experiments with document-level machine translation. The data sets have been used in previous studies and provided here for replicability and comparison with other systems. The data sets are taken from the English-German news translation task at WMT 2019 and the English-German bitext in the OpenSubtitles collection v2016 from OPUS. All data sets are sentence aligned with corresponding lines being aligned to each other. Document boundaries are marked with empty lines (on both sides of the parallel corpus). The data set has been used in the following publication: @inproceedings{scherrer-tiedemann-loaiciga-2019, title = "Analysing concatenation approaches to document-level NMT in two different domains", author = {Scherrer, Yves and Tiedemann, J{\"o}rg and Lo{\'a}iciga, Sharid}, booktitle = "Proceedings of the Third Workshop on Discourse in Machine Translation", month = nov, year = "2019", address = "Hong-Kong", publisher = "Association for Computational Linguistics", } Please, cite that paper if you use the data set in your own work.

This dataset contains a proof-of-concept GIS database of over 29,000 individual historical road polyline segments as a shapefile dataset, covering over 11,000 km2 in the western Finland from the city of Turku to northern parts of the province of Satakunta. These polylines capture the regional layout of the overland transport infrastructure of late nineteenth and early twentieth century Finland. {"references": ["Oksanen, Eljas, Saarenp\u00e4\u00e4, Ida. and Lahtinen, Anu. 2023. The HISCOM Project. Exploring Methodologies for Large-scale Digitisation of Historical Roadways. SKAS 2/2022, 8-14"]}

This dataset merges nitrogen data from the Yedoma domain. It includes numerous fieldwork campaigns, which take place since 1998. In total 467 samples from the active layer (seasonally thawed layer), 175 samples from perennially frozen Holocene cover deposits, 479 samples from thermokarst deposits in drained thermokarst, 175 in-situ thawed, diagenetically (anaerobic microbial decomposition possible during unfrozen phase) altered Yedoma deposits (called Taberite), and 917 samples from frozen Yedoma deposits are included. Moreover it includes a NH4+ and NO3- quantification basing on of 658 samples, including 378 data points for NH4+ (active layer, 93; Holocene cover, 108; thermokarst sediment, 138; Taberite, 0; Yedoma deposit, 39) and 542 data points for NO3- (active layer, 94; Holocene cover, 137; thermokarst sediment, 119; Taberite, 6; Yedoma deposit, 186). The bootstrapping code we adjusted for this study is available from Zenodo (Jongejans & Strauss, 2020, doi:10.5281/zenodo.3734247). The code is published under a GNU General Public License v3.0. The included areal estimation of the Yedoma domain was used from the IRYP database (Strauss et al., 2022, doi:10.1594/PANGAEA.940078).