|
Download README.md from miserableface/opensubtitle-melayu-examples: direct link, hf CLI and curl.
- Browser
- Download file 1.95 kB
-
https://huggingface.co/datasets/miserableface/opensubtitle-melayu-examples/resolve/main/README.md
- Command line
-
hf download hf://datasets/miserableface/opensubtitle-melayu-examples/README.md
-
curl -L -o README.md https://huggingface.co/datasets/miserableface/opensubtitle-melayu-examples/resolve/main/README.md
1.95 kB
| license: unknown | |
| language: | |
| - ms | |
| - id | |
| source_datasets: | |
| - opus | |
| - opensubtitles | |
| tags: | |
| - dictionary | |
| - malay | |
| - indonesian | |
| - example-sentences | |
| - opus | |
| size_categories: | |
| - 100K<n<1M | |
| # Melayu OPUS OpenSubtitles examples | |
| Ms-en and id-en example sentences mined from [OpenSubtitles v2024](https://opus.nlpl.eu/OpenSubtitles/en&ms/v2024/OpenSubtitles) (17,322,086 raw pairs scanned), matched against the melayu headword inventory via the shared entry_id join contract, and capped at the 20 best unique sentences per headword. | |
| ## Columns | |
| - `entry_id` (string): headword id via the melayu-entries join contract | |
| - `word` (string): the matched headword (ms or id) | |
| - `lang` (string): ms or id | |
| - `pos` (string): pos from the melayu-entries join contract | |
| - `orig` (string): the ms/id subtitle line | |
| - `en` (string): aligned English line | |
| - `rank` (int32): OpenSubtitles frequency rank (0 = outside top-50k) | |
| - `matched_form` (string): exact | suffixed (plugin-less morphology pass deferred) | |
| - `source` (string): OPUS zip URL | |
| ## Provenance & attribution | |
| - OPUS OpenSubtitles v2024 moses zip: https://object.pouta.csc.fi/OPUS-OpenSubtitles/v2024/moses/en-ms.txt.zip | |
| - extracted: 2026-10-27 | |
| - headword coverage: ms 6,649 / id 5,051 words | |
| **Important (OPUS/OpenSubtitles request): if you use this corpus, add a link to [http://www.opensubtitles.org/](http://www.opensubtitles.org/) to your website and reports/publications produced with the data.** | |
| Cite: P. Lison and J. Tiedemann, 2016, *OpenSubtitles2016: Extracting Large Parallel Corpora from Movie and TV Subtitles*, LREC 2016. | |
| Due diligence text (as included in the OPUS package): 'We do not own any of the text from which the data has been extracted. We only offer files that we believe we are free to redistribute.' | |
| ## License | |
| Upstream subtitle text is community-contributed; this dataset follows OPUS's redistribution practice (CC-BY-style attribution). See the upstream LICENSE shipped in the moses zip. |