I aimed to use machine learning for airline sentiment analysis. The project involved exploratory data analysis, data balancing, and implementing Naive Bayes and Decision Tree models. The Naive Bayes model, particularly with TF-IDF features and oversampling or undersampling, outperformed the Decision Tree, achieving 0.78 accuracy and 0.71 F1 score. The project highlighted the importance of data preprocessing, feature engineering, and model selection in sentiment analysis tasks.