Hệ thống phân tích cảm xúc tiên tiến cho dịch vụ công Việt Nam | Advanced sentiment analysis for Vietnamese public services
- 🎯 State-of-the-Art Performance: 88.8% overall accuracy with outstanding 98.3% negative sentiment detection
- 🔥 Advanced NLP: Powered by PhoBERT - the leading Vietnamese language model
- ⚡ Real-time Analysis: Process and analyze feedback instantly
- 🎨 Multi-aspect Analysis: Detect multiple sentiments within single reviews
- 📊 Comprehensive Dataset: Trained on 85,500+ authentic public service reviews
- 🛠️ Production-Ready: Scalable architecture for enterprise deployment
Traditional sentiment analysis tools struggle with Vietnamese public service reviews due to:
- Complex administrative terminology
- Regional linguistic variations
- Mixed formal/informal language
- Context-dependent expressions
Our system specifically addresses these challenges, providing government agencies with accurate, actionable insights from citizen feedback.
- 📝 Specialized Vietnamese text preprocessing
- 🔄 Handles code-switching (Vietnamese-English)
- 🎯 Domain-specific terminology recognition
- 🔧 Regional language variation support
- 🧠 Fine-tuned PhoBERT model
- 📈 Context-aware sentiment detection
- 🎭 Multi-aspect sentiment classification
- ⚡ Real-time processing capabilities
- 🚀 Scalable architecture
- 📊 Comprehensive analytics dashboard
- 🔄 Continuous monitoring
- 📱 Multi-platform support (Google Play, Google Maps)
| Metric | Score |
|---|---|
| Overall Accuracy | 88.8% |
| Negative Sentiment Detection | 98.3% |
| Positive Sentiment Detection | 79.0% |
| Processing Speed | Real-time |
| Scale | 85,500+ reviews analyzed |
- Core: Python 3.8+, PyTorch
- Model: PhoBERT (fine-tuned)
- Processing: Transformers, Vietnamese NLP tools
- Analysis: Custom sentiment analysis pipeline
- Deployment: Docker-ready
- 📊 Real-time service quality monitoring
- 🎯 Early issue detection
- 📈 Trend analysis
- 🔄 Service improvement tracking
- 🧪 Benchmark dataset for Vietnamese NLP
- 📚 Public service sentiment analysis
- 🔬 E-government feedback analysis
# Clone the repository
git clone https://github.com/YCN-AFS/GovService_SentimentAI.git
# Install dependencies
pip install -r requirements.txt
# Run the analysis
python analyze.py --input your_data.csvComprehensive documentation available at docs/README.md
We welcome contributions! See CONTRIBUTING.md for guidelines.
If you use this work in your research, please cite:
@article{phobert-public-service-2024,
title={Sentiment Analysis System for Public Service Reviews Using PhoBERT},
author={[Authors]},
journal={[Journal]},
year={2024}
}[License information to be added]
⭐ Don't forget to star this repo if you find it useful!