This project compares the performance of RoBERTa (a transformer-based deep learning model) and VADER (a lexicon-based sentiment analysis tool) in analyzing sentiment from text data.
- VADER: Lexicon-based sentiment analysis (faster, rule-based)
- RoBERTa: Deep learning-based sentiment analysis.
- Performance comparison on different text inputs.
-
Install Dependencies. pip install torch transformers nltk
-
Download NLTK resources (for VADER). import nltk nltk.download('vader_lexicon')
VADER: Uses predefined sentiment scores, making it faster but sometimes less accurate for complex language. RoBERTa: Uses deep learning, providing context-aware sentiment classification (Positive, Neutral, Negative).
Feel free to fork and improve this project by adding more evaluation metrics or datasets!
MIT License