This project performs sentiment analysis on user-generated content from Reddit, Twitter, and YouTube using Natural Language Processing (NLP) and Machine Learning.
- reddit-sentiment-analysis: Analyzes sentiment in Reddit comments
- tweet-sentiment-analysis: Analyzes sentiment in tweets
- youtube-sentiment-analysis: Analyzes sentiment in YouTube comments
- Fetches real-time data from social media platforms
- Cleans and preprocesses text (tokenization, stopword removal, lemmatization)
- Uses TF-IDF and Word Embeddings for feature extraction
- Trains ML models (Logistic Regression, SVM, Naive Bayes) for sentiment classification
- Generates interactive visualizations of sentiment trends
Results & Outputs • Processed Data: data/processed/ • Raw Data: data/raw/ • Trained Models: models/ • Visualizations: visualization/
Future Improvements • Real-time sentiment tracking across platforms • Integration with deep learning models (LSTM, BERT) • Multilingual support for sentiment analysis