# Social Media Content Moderation

Multilingual toxicity classifier based on xlm-roberta-large architecture, supporting toxicity detection in 9 languages

A Spanish cyberbullying detection model fine-tuned from RoBERTa-base-bne, specifically designed for social media content in the Colombian region

A BERT-based multilabel hate speech classifier designed to identify categories of hate speech in text, including race, religion, origin, gender, sexual orientation, age, and disability.

A text classification model for detecting and categorizing hate speech.

A Russian toxicity text named entity recognition model fine-tuned based on ruBERT-tiny, capable of identifying toxic content in text

This model is a branch version based on the bert-based-uncased-hatespeech-movies model, used to classify text as normal, offensive, or hate speech.

A multilingual text toxicity classification model trained on the JIGSAW Toxic Comment Classification Challenge dataset, supporting 10 languages.

Fine-tuned Russian text toxicity multi-label classification model based on rubert-tiny

Binary hate speech detection model fine-tuned on the Hindi HASOC hate speech dataset 2021

This model is used for detecting hate speech in Spanish, fine-tuned based on the multilingual BERT model.

This model is fine-tuned based on multilingual BERT, specifically designed for detecting hate speech in German, using monolingual (English) training data.

This model is used to detect hate speech in Arabic, fine-tuned based on multilingual BERT, with the best validation score of 0.877609

This model is used to detect offensive content in Tamil code-mixed text, trained based on the XLM-Roberta-Base model, and performed excellently in the EACL 2021 Dravidian Language Offensive Language Identification Shared Task.

Multilingual BERT fine-tuned model for Italian hate speech detection, trained using English data

This model is designed for detecting hate speech in French, fine-tuned from a multilingual BERT model using a monolingual training setup.