---
task_categories:
- text-generation
language:
- ru
tags:
- history
- linguistics
- medicine
- russian
- keywords
- keyphrases
size_categories:
- 1K<n<10K
---

A dataset for extracting or generating keywords for scientific texts in Russian. It contains annotations of scientific articles from four domains: mathematics and computer science (a fragment of <a href="https://data.mendeley.com/datasets/dv3j9wc59v/1">the Keyphrases CS&Math Russian dataset</a>), history, linguistics, and medicine. For more details about the dataset please refer the original paper: https://arxiv.org/abs/2409.10640

## Dataset Structure

<ul>
  <li><b>abstract</b>, an abstract in a string format;</li>
  <li><b>keywords</b>, a list of keyphrases provided by the authors of the paper.</li>
</ul>


![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/63776c8980c7d5df289aced7/EFWQtY-y8lkymDcAQbCsO.jpeg)

## Source Data

cyberleninka.ru, mathnet.ru

## Citation

```
@article{glazkova2024exploring,
  title={Exploring Fine-tuned Generative Models for Keyphrase Selection: A Case Study for Russian},
  author={Glazkova, Anna and Morozov, Dmitry},
  journal={arXiv preprint arXiv:2409.10640},
  year={2024}
}
```