This Jupyter notebook performs sentiment analysis on Twitter data. The project includes data preprocessing, training a machine learning model, and evaluating its performance.
The main features of the repository include:
- Data Preprocessing: Cleaning and stemming of tweet text.
- Model Training: Training a logistic regression model to classify tweets as positive or negative.
- Model Evaluation: Assessing the model's accuracy on test data.
- Model Saving: Saving the trained model for future use.
- Prediction: Using the saved model to predict the sentiment of new tweets.
Technologies used: Python, Pandas, Scikit-learn, and Pickle.