Classifier of opinion conveyed by vaccine-related content in Italian language
A monolingual model for classifying the opinion conveyed through vaccine-related content in Italian language. The model was trained on 36,722 and independently tested on 9,299 social media content between Facebook posts, Twitter tweets, Instagram media and YouTube videos. It is a fine-tuned version of bert-base-multilingual-cased.
Model output
The model classifies each input into one of three distinct classes:
- Anti-vax
- Neutral
- Pro-vax
Citation info and BibTeX entry
https://arxiv.org/abs/2207.12264
@article{Bru2022,
title={Dynamics of (mis)information flow and engaging power of narratives},
author={Brugnoli, Emanuele and Delmastro, Marco},
journal={ArXiv},
year={2022},
volume={abs/2207.12264}
}
https://arxiv.org/abs/2301.05961
@article{Gal2023,
title={Unveiling the Hidden Agenda: Biases in News Reporting and Consumption},
author={Galeazzi, Alessandro and Peruzzi, Antonio and Brugnoli, Emanuele and Delmastro, Marco and Zollo, Fabiana},
journal={ArXiv},
year={2023},
volume={abs/2301.05961}
}