Analysis of Long Short-Term Memory (LSTM) and Extreme Gradient Boosting (XGBoost) Algorithms to Predict the Number of Airplane Passengers at Makassar Sultan Hasanuddin International Airport : Systematic Literature Review

Ainul Idham
Efy Yosrita

Jurnal E-Komtek
Vol 9 No 1 (2025)

This study compares the performance of Long Short-Term Memory (LSTM), Extreme Gradient Boosting (XGBoost), and hybrid techniques to forecast the number of aircraft passengers. This analysis was carried out utilizing the Systematic Literature Review (SLR) method and the PRISMA approach. Only 11 of the 44,564 items filtered during the initial round met the inclusion requirements. The LSTM model performed well in capturing time series patterns, however XGBoost was more robust when employed on data with noise and outliers. The hybrid model (LSTM + XGBoost) performed the best, with an average accuracy of 96%, RMSE of 0.015, and MAPE of 2.45%. This demonstrates that the hybrid technique is quite good in predicting the number of airplane passengers, particularly for complicated, dynamic, and seasonal time series data. These findings are recommended for the development of machine learning-based prediction systems for airports.