This is an emotion classification model based on further pre-training of BERTweet-large with preferential masking of emotion words and fine-tuning on a subset of a self-labeled emotion dataset (Lykousas et al., 2019) that corresponds to Anger, Fear, Sadness, Joy, and Affection. The paper, LEIA: Linguistic Embeddings for the Identification of Affect provides further details on the model and its evauation.
See LEIA-base for a similar model based on BERTweet-base.
Citation
Please cite the following paper if you find the model useful for your work:
@article{aroyehun2023leia,
title={LEIA: Linguistic Embeddings for the Identification of Affect},
author={Aroyehun, Segun Taofeek and Malik, Lukas and Metzler, Hannah and Haimerl, Nikolas and Di Natale, Anna and Garcia, David},
journal={EPJ Data Science},
volume={12},
year={2023},
publisher={Springer}
}