bengali-sentiment-analysis

🏆 Bengali Sentiment Analysis

A production-ready Bengali sentiment analysis system using XLM-RoBERTa for analyzing Bengali, English, and code-mixed text sentiment.

🎯 Features

• 🚀 Immediate Inference - Pre-trained model ready for immediate use
• 💬 Bengali/English Support - Analyzes Bengali, English, and code-mixed text
• 📊 High Accuracy - ~80% accuracy on benchmark datasets
• 🌐 Multi-language - Supports Bengali (বাংলা) and English
• 🔧 Production Ready - Complete API, CLI, and web deployment options

📦 Installation

pip install transformers torch

🚀 Quick Start

from transformers import pipeline

# Load the model
analyzer = pipeline(
    "sentiment-analysis",
    model="OsamaBinLikhon/bengali-sentiment-analysis"
)

# Analyze Bengali text
result = analyzer("আমি ভালো আছি")
print(result)
# [{'label': 'positive', 'score': 0.85}]

# Analyze English text
result = analyzer("This is amazing!")
print(result)
# [{'label': 'positive', 'score': 0.94}]

📊 Supported Sentiments

📁 Repository Structure

OsamaBinLikhon/bengali-sentiment-analysis/
├── README.md           # This file
├── model_card.md       # Detailed model documentation
├── config.json         # Model configuration
├── pytorch_model.bin   # Model weights
├── tokenizer.json      # Tokenizer files
├── vocab.txt           # Vocabulary
└── special_tokens_map.json

🔧 Advanced Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

# Load model and tokenizer
model_name = "OsamaBinLikhon/bengali-sentiment-analysis"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

# Analyze with custom settings
text = "আমি খুব খুশি"
inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=128)
outputs = model(**inputs)
predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
print(predictions)

📈 Performance

• Accuracy: ~80%
• Macro F1: ~0.80
• Inference Speed: < 100ms per sample on CPU

🌐 Language Support

• ✅ Bengali (বাংলা)
• ✅ English
• ✅ Code-mixed Bengali-English
• 🔄 Other Indic languages (experimental)

📝 Example Outputs

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Hugging Face for the transformers library
• Cardiff NLP for the base model
• The Bengali NLP community for their support

---

Made with ❤️ by Bengali AI Team