psytechlab/RuSentiTweet

---
dataset_info:
features:
- name: text
dtype: string
- name: label
dtype: string
- name: id
dtype: int64
splits:
- name: train
num_bytes: 1148348
num_examples: 9641
- name: test
num_bytes: 317153
num_examples: 2679
- name: val
num_bytes: 125692
num_examples: 1072
download_size: 1048892
dataset_size: 1591193
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: val
path: data/val-*
task_categories:
- text-classification
language:
- ru
tags:
- russian
- sentiment_analysis
- social_media
size_categories:
- 1K<n<10K
---

## Disclaimer

This is a reupload of the dataset, which is originally stored on [GitHub](https://github.com/sismetanin/rusentitweet). We only added a `val` split that is 10% of the original `train` split for a convenience. Next the original description is following.

# RuSentiTweet: A Sentiment Analysis Dataset of General Domain Tweets in Russian

This repository contains RuSentiTweet, a sentiment analysis dataset of 13,392 general domain tweets in Russian, which were created within the paper ["RuSentiTweet: A Sentiment Analysis Dataset of General Domain Tweets in Russian"](https://doi.org/10.7717/peerj-cs.1039). RuSentiTweet was manually annotated (moderate inter-rater agreement) using [RuSentiment](https://aclanthology.org/C18-1064/) guidelines into 5 classes: Positive, Neutral, Negative, Speech Act, and Skip. As a source of data, we used [Twitter Stream Grab](https://archive.org/details/twitterstream), a historical collection of tweets obtained from the general Twitter API stream.

Citation:
```
@article{smetanin2022rusetitweet,
title = {RuSentiTweet: A Sentiment Analysis Dataset of General Domain Tweets in Russian},
author = {Sergey Smetanin},
journal = {PeerJ Computer Science},
volume = {8},
pages = {e1039},
year = {2022},
doi = {10.7717/peerj-cs.1039},
publisher = {PeerJ Inc.}
}
```