In this project, we compared Spanish BERT and Multilingual BERT in the Sentiment Analysis task.

All our code for the DEtection of TOXicity in comments In Spanish (DETOXIS) at IberLEF 2021.

This repository is the implementation of CaTrBETO: Spanish Pre-Trained Model for Sentiment Classification in Twitter

BETO formalizes the ignorance of an AI. Materializes raw semantic intent into fully traceable, auditable software.

Count metric syllables and generate décimas based on BETO model.

Thesis

A fine-tuned model for detecting hate speech in Spanish, targeting racism, homophobia, sexism, and transphobia. Built on dccuchile/bert-base-spanish-wwm-cased, it’s trained on the manueltonneau/spanish-hate-speech-superset dataset with additional Paul dataset samples for improved accuracy and nuance. Ideal for content moderation.

Transformers applied to Stance Detection for Spanish and Catalan languages

Transformer-based sexism detection in Spanish tweets

NLP related code.

This API is focused on providing the service of analyzing any text and identifying whether it is consensual sexting or not.

Contains the notebook used during the shared task SMM4H-ProfNER

Trabajo de fin de máster en Inteligencia Artificial de la UNIR