This model is a Spanish-language fake news detector based on bert-base-multilingual-cased. It was fine-tuned on a custom dataset of real and fake news articles, combining titles and body text for contextual understanding.

The model performs binary text classification, predicting whether a news item is likely to be Real or Fake. It achieves high accuracy and balanced performance across precision, recall, and F1 score, making it suitable for public inference and deployment.

Key features:

Language: Spanish 🇪🇸
Task: Text Classification (Fake vs Real)
Base model: BERT Multilingual
Training framework: Hugging Face Transformers
Evaluation metrics: Accuracy 95.47%, F1 Score 95.44%