Machine Learning

Airline Customer Sentiment Analysis Using Machine Learning

The Airline Customer Sentiment Analysis Using Machine Learning system is a comprehensive machine learning and NLP solution that analyzes airline-related tweets to classify customer sentiment into negative, neutral, or positive categories. This project is ideal for B.Tech, MCA, CSE, and engineering students as their final year project. It uses advanced ensemble algorithms like XGBoost and Random Forest with TF-IDF vectorization to achieve 82.34% accuracy.

The system leverages text preprocessing, feature engineering (TF-IDF vectorization, sentiment word scores, confidence metrics), and leak-free validation using SMOTE within cross-validation folds. It provides interactive visualizations, feature importance analysis, model comparison, and real-time sentiment prediction to help airlines monitor brand perception and improve customer satisfaction.

• Multi-Class Sentiment Classification
• XGBoost (82.34% Accuracy)
• Random Forest (79.87% Accuracy)
• TF-IDF Text Feature Extraction
• SMOTE for Class Imbalance
• Leak-Free 5-Fold CV Validation
• Feature Importance Analysis
• Real-time Sentiment Prediction
• Interactive Visualizations
• Flask Web Application