Sentiment Sage is a REST API built using Docker, incorporating a machine learning model reliant on NLTK (Natural Language Toolkit). This API serves as a standalone sentiment prediction system. It analyzes input sentences and produces sentiment scores, considering positivity, negativity, and neutrality. The underlying sentiment classification strategy is rule-based, employing VADER (Valence Aware Dictionary and Sentiment Reasoner), which is adept at interpreting sentiments found in social media.
- Architecture: The system is containerized using Docker for easy deployment and management.
- Machine Learning Model: Built on NLTK, the model processes text input to predict sentiment.
- Functionality: The API receives sentences as input and returns sentiment scores.
- Sentiment Scoring: Each sentence is analyzed for positivity, negativity, and neutrality.
- Sentiment Classification: VADER is utilized for sentiment classification.
- VADER: VADER is a lexicon and rule-based sentiment analysis tool, specialized for social media sentiments.
- Rule-Based Strategy: Sentiment classification is driven by predetermined rules within VADER.
- Output: The API provides a comprehensive sentiment analysis report for each input sentence.
sequenceDiagram
Web Server -> Model:
loop Prediction Request with params
Web Server --> Model: Response containing predictions
end
docker build -t sentiment-sage .
docker run -p 8000:8000 sentiment-sage