Sentiment Analysis Using PhoBERT

Harnessing PhoBERT and the UIT-VSFC dataset for advanced Vietnamese sentiment analysis.

📑 Overview

This project leverages PhoBERT with the UIT-VSFC dataset to classify sentiment in Vietnamese text. By utilizing the power of a transformer-based model fine-tuned for Vietnamese, this project achieves highly accurate sentiment predictions, ideal for natural language understanding applications in Vietnamese.

✨ Key Features

• State-of-the-Art NLP: Employs PhoBERT, a leading model for Vietnamese language understanding.
• UIT-VSFC Dataset: Trained on a well-curated dataset with extensive Vietnamese sentiment data.
• Robust Sentiment Analysis: Classifies text sentiment as positive, negative, or neutral with precision.

🔧 Installation

1. Clone the repository:

   git clone https://github.com/LeHuyHongNhat/Sentiment-Analysis-Using-PhoBERT.git

2. Navigate to the project directory:

   cd Sentiment-Analysis-Using-PhoBERT

📝 Usage

Data Preparation

Download the UIT-VSFC dataset in .

Model Training

Fine-tune PhoBERT on the UIT-VSFC dataset

Evaluation

Evaluate model performance on the test dataset

📊 Results

The fine-tuned model achieves high accuracy on the UIT-VSFC dataset, demonstrating effectiveness for sentiment analysis tasks in Vietnamese. Detailed metrics can be found in the notebooks.

🚀 Future Work

• Dataset Expansion: Incorporate additional Vietnamese sentiment datasets.
• Multi-Label Sentiment: Extend the model to capture more nuanced sentiment variations.

📄 License

This project is licensed under the MIT License.

🤝 Acknowledgments

• PhoBERT: PhoBERT on Hugging Face
• UIT-VSFC Dataset: UIT-VSFC Sentiment Dataset

Made with ❤️ by Lê Huy Hồng Nhật