metadata
license: mit
task_categories:
- text-classification
language:
- tr
size_categories:
- 1K<n<10K
pretty_name: mide22-tr
MiDe22: An Annotated Multi-Event Tweet Dataset for Misinformation Detection
5,064 Turkish tweets with their misinformation labels for several recent events between 2020 and 2022, including the Russia-Ukraine war, COVID-19 pandemic, and Refugees. The dataset includes user engagements with the tweets in terms of likes, replies, retweets, and quotes. For user engagements please use contact at the end of the dataset card.
Data Fields
tweet: astringfeature.label: a classification label, with possible values includingTrue,False,Other.
Data Size
| classes | true | false | other |
|---|---|---|---|
| tweets | 669 | 1,732 | 2,663 |
Annotations
Tweets are labeled in terms of three classes: True, False, and Other.
True: tweets with the correct information regarding the corresponding event.False: tweets with misinformation on the corresponding event.Other: tweets that cannot be categorized under false or true information.
Annotation process
This dataset is annotated by five annotators. Each tweet is annotated by at least two annotators, we calculate Krippendorf’s alpha reliability to measure interannotator agreement (IAA). The resulting alpha coefficient is 0.791. Details are given in our paper.
Dataset Sources
- Repository: More details on GitHub
- Paper: LREC-COLING 2024
Citation
@inproceedings{toraman-etal-2024-mide22-annotated,
title = "{M}i{D}e22: An Annotated Multi-Event Tweet Dataset for Misinformation Detection",
author = "Toraman, Cagri and
Ozcelik, Oguzhan and
Sahinuc, Furkan and
Can, Fazli",
booktitle = "Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)",
month = may,
year = "2024",
address = "Torino, Italia",
publisher = "ELRA and ICCL",
url = "https://aclanthology.org/2024.lrec-main.986",
pages = "11283--11295",
}
Dataset Card Contact
ogozcelik[at]gmail[dot]com