The corpus includes 4,037 human-labelled Arabic Twitter messages for four high-risk flood events that occurred in 2018 The tweets labelled based on relatedness to the crisis and information type.

If you use the dataset in your research, please cite the following paper:

Alharbi, A. and Lee, M., 2019. Crisis Detection from Arabic Tweets. In Proceedings of the 3rd Workshop on Arabic Corpus Linguistics (pp. 72-79).

BibTeX:

@inproceedings{alharbi2019crisis,
 title={Crisis Detection from Arabic Tweets},
 author={Alharbi, Alaa and Lee, Mark},
 booktitle={Proceedings of the 3rd Workshop on Arabic Corpus Linguistics},
 pages={72--79}, 
 year={2019}
}