Revolutionizing Airline Customer Satisfaction Analysis with Machine Learning Techniques

The customer’s experience is one of the most critical concerns for the airline industry. This study investigates customer satisfaction in the airline industry through machine learning analysis, employing decision trees, random forests, support vector machines, and XGBoost models to predict customer satisfaction levels. Each model offers a unique approach to analyzing and predicting customer satisfaction levels, contributing to a comprehensive evaluation of their effectiveness. The data undergoes exploratory data analysis (EDA) and preprocessing, enabling quality assessment. XGBoost consistently outperforms other models in predictive accuracy. Key metrics, including accuracy, precision, recall, F1 score, and ROC AUC, are employed to evaluate model performance. The XGBoost model was identified as being the best among the four models tested (accuracy: 0.9576, precision: 0.9683, recall: 0.9528, F1 score: 0.9574, ROC AUC score: 0.9940). These findings contribute valuable insights into improving airline customer satisfaction and inform decision-making processes within the industry.