A project to classify input text as hate speech or not hate speech using an LSTM (Long Short-Term Memory) neural network.
Dataset
Trained on:
- Hate Speech and Offensive Language dataset
- Twitter hate speech dataset from Kaggle
Model
An LSTM-based recurrent neural network which is well-suited for sequential text data, capturing long-range dependencies in language.
Tech Stack
| Component | Technology |
|---|---|
| Model | LSTM (TensorFlow / Keras) |
| NLP Preprocessing | Python (tokenization, padding) |
| Frontend | Gradio |
| Deployment | HuggingFace Spaces |