Datasets:

Modalities:
Text
Formats:
csv
Size:
< 1K
Libraries:
Datasets
pandas
License:
ErykWdowiak's picture
Update README.md
7f3da3c verified
metadata
license: cc-by-sa-4.0
language:
  - it
  - scn

Good Sicilian from WikiMatrix

"Good Sicilian" presents an 800-year literary tradition. "Good Sicilian" is the literary language described in the three grammar textbooks that Arba Sicula has published.

Of those three, the first by Kirk Bonner (2001) strongly influenced Sicilian Wikipedia's style guide. The two more recent grammars by Gaetano Cipolla, Learn Sicilian (2013) and Learn Sicilian II (2021), present the same Sicilian literary language, but were published several years after Sicilian Wikipedia began its work.

Some "Good Sicilian" from WikiMatrix

With Arba Sicula's grammar textbooks at hand, Project Napizia identified 514 Italian-Sicilian translation pairs from the WikiMatrix project and edited them to perfection.

We hope researchers and practitioners will use this WikiMatrix data to help language models learn "Good Sicilian." And we'll update this project with more public collections of "Good Sicilian" at next opportunity.

Dataset Card -- Italian-Sicilian from WikiMatrix

Dataset Summary

This dataset contains 514 human-edited Italian-Sicilian pairs from the WikiMatrix project.

Licensing Information

Facebook Research distributed the original mined data under the Creative Commons Attribution-ShareAlike license (CC BY-SA). By using this, you are also bound to the respective Terms of Use and License of the original source.

Source

Schwenk et al (2021). "WikiMatrix: Mining 135M Parallel Sentences in 1620 Language Pairs from Wikipedia."