Datasets:
Tasks:
Token Classification
Modalities:
Text
Formats:
json
Languages:
Ukrainian
Size:
10K - 100K
DOI:
task_categories: | |
- token-classification | |
language: | |
- uk | |
pretty_name: 'Silver Ukrainian Coreference Dataset ' | |
tags: | |
- coreference-resolution | |
- coreference | |
- anaphora | |
size_categories: | |
- 10K<n<100K | |
# Silver Ukrainian Coreference Dataset | |
## Dataset Description | |
### Dataset Summary | |
A silver coreference resolution dataset for the Ukrainian language. The dataset was generated automatically with the usage of the word alignment method from the following English dataset: https://github.com/d5555/Coreference-dataset. | |
The word alignment method was implemented by Andrii Kursin (aqrsn@ukr.net). | |
### Languages | |
- Ukrainian | |
## Dataset Structure | |
### Data Fields | |
Each sample of the dataset consists of the following fields: | |
- **doc_key** - document identifier. | |
- **clusters** - list of clusters, where each cluster consists of the list of mentions. Each mention is represented as a list of two indices: the first index denotes the first word of the mention, the second index denotes the last word of the mention. | |
- **sentences** - list of sentences where each sentence is represented as a list of words. | |
- **tokens** - list of words. | |
- **speakers** - list of speakers which is currently filled with dummy input. | |
### Data Splits | |
The dataset is divided into two parts: | |
- training set; | |
- validation set. | |
A test set is absent as far as the dataset is generated automatically. | |
## Dataset Creation | |
### Source Data | |
The dataset was created from the following dataset: https://github.com/d5555/Coreference-dataset. | |
### Contributions | |
The code for the translation of samples with further alignment was created by Andrii Kursin (aqrsn@ukr.net). The dataset was generated by Artem Kramov (https://www.linkedin.com/in/artem-kramov-0b3731100/). |