This project goal is to implement, test and compare several different deep learning models w.r.t. the HaSpeeDe (i.e., hate-speech detection) task from the Evalita 2018 challenge.
The dataset used for development and testing was obtained from Evalita 2018. An overview of the task, as well as the results' of previous participants can be found at HaSpeeDe-1 overview [1]
More information about the considered models, as well as the methodology we employed in our analyses and our results can be found in the report.
[1] Bosco, C., Felice, D. O., Poletto, F., Sanguinetti, M., & Maurizio, T. (2018). Overview of the evalita 2018 hate speech detection task. In Ceur workshop proceedings (Vol. 2263, pp. 1-9). CEUR.