Kaggle
CC BY-SA 4.0
DOI 10.34740/kaggle/dsv/13468825
Bagdhara — Early Public Release
The first public release of the Bagdhara idiom corpus, and the most downloaded Bengali idiom resource on Kaggle.
What it contains
What this is
The first public release of the Bagdhara corpus: 10,361 Bengali idioms, one JSON file per idiom. It went out before the expert-validated Bangla Bagdhara release that accompanies the LREC 2026 paper, and it is still the most downloaded Bengali idiom resource on Kaggle.
| Idioms | 10,361 (one JSON file each) |
| Downloads | 5,226 |
| Views | 1,069 |
| Size | 16.4 MB |
| Licence | CC BY-SA 4.0 |
Which version should I use?
| If you need | Use |
|---|---|
| To reproduce work that cited this release | This dataset, DOI 10.34740/kaggle/dsv/13468825 |
| The expert-validated corpus and LREC 2026 benchmark | Bangla Bagdhara |
This release stays online so that everyone who built on it early can still cite and reproduce their work.
Downloads
5,226
Kaggle usability
10.0
Superseded by
Bangla Bagdhara