license: cc-by-nc-sa-4.0
SciPar Parallel Documents
Dataset Description
This dataset contains parallel documents (i.e., titles & abstracts) extracted from academic theses, dissertations, and other scientific texts.
In the original paper, we've extracted 9.17M sentence pairs in 31 language pairs from 86 repositories.
This version has been created through further processing and filtering to extract parallel documents instead of parallel sentences.
To do this, we kept only the parallel titles and abstracts of academic records with high (mean) alignment scores by also using other filters (e.g., word ratio, non-empty abstracts, etc.).
Please note that this dataset will keep getting updated to include more language pairs, per the original paper.
Table 1: Number of parallel documents per language pair
| Lang. Pair | # Parallel Docs |
|---|---|
| EN-DE | 57,387 |
| EN-EL | 55,833 |
| EN-ES | 25,844 |
| EN-FR | 130,750 |
| EN-IT | 3,860 |
| TOTAL | 273,674 |
Related Datasets
The original datasets can be found in ELRC-SHARE:
Citation
@inproceedings{roussis2022scipar,
title={SciPar: A collection of parallel corpora from scientific abstracts},
author={Roussis, Dimitrios and Papavassiliou, Vassilis and Prokopidis, Prokopis and Piperidis, Stelios and Katsouros, Vassilis},
booktitle={Proceedings of the Thirteenth Language Resources and Evaluation Conference},
pages={2652--2657},
year={2022}
}