Off_MorRoBERTa
Off_MorRoBERTa is a Transformer-based language model designed to classify comments in the Moroccan dialect as abusive or non-abusive.
About Off_MorRoBERTa
Off_MorRoBERTa, developed specifically for the Moroccan dialect, provides an effective approach for detecting abusive content. The model is based on MorRoBERTa, which was fine-tuned on a labeled dataset created for hate speech detection in Moroccan dialect texts.
The dataset used for training was built by aggregating and preprocessing data from publicly available sources, including MDMD, OMCD, and MDOLDD.
Usage
The model weights can be loaded using transformers library by HuggingFace.
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("otmangi/Offensive_Darija_MorRoBERTa")
model = AutoModel.from_pretrained("otmangi/Offensive_Darija_MorRoBERTa")
Related Paper
For more details about the dataset, methodology, and evaluation, please refer to our paper:
https://accentsjournals.org/paperinfo.php?journalPaperId=1809Contact
For any inquiries, feedback, or requests, please feel free to reach out to :
