Comparative Analysis of Optimizer Effectiveness in GRU and CNN-GRU Models for Airport Traffic Prediction

Riyadi, Willy and Jasmir, Jasmir (2024) Comparative Analysis of Optimizer Effectiveness in GRU and CNN-GRU Models for Airport Traffic Prediction. Jurnal Ilmiah Teknik Elektro Komputer dan Informatika (JITEKI), 10 (3). pp. 580-593.

Text 7-Comparative Analysis of Optimizer Effectiveness in Gru and Cnn-Gru Models for Airport Traffic Prediction.pdf Download (1MB)

Abstract

The COVID-19 pandemic has posed significant challenges to airport traffic management, necessitating accurate predictive models. This research evaluates the effectiveness of various optimizers in enhancing airport traffic prediction using Deep Learning models, specifically Gated Recurrent Units (GRU) and Convolutional Neural Network-Gated Recurrent Units (CNN-GRU). We compare the performance of optimizers including RMSprop, Adam, Nadam, AdamW, Adamax, and Lion, and analyze the impact of their parameter tuning on model accuracy. Time series data from airports in the United States, Canada, Chile, and Australia were used, with preprocessing steps like filtering, cleaning, and applying a MinMax Scaler. The data was split into 80% for training and 20% for testing. Our findings reveal that the Adam optimizer paired with the GRU model achieved the lowest Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) in the USA. The study underscores the importance of selecting and tuning optimizers, with ReduceLROnPlateau used to adjust the learning rate dynamically, preventing overfitting and improving model convergence. However, limitations include dataset imbalance and region-specific results, which may affect the generalizability of the findings. Future research should address these limitations by developing balanced datasets and exploring optimizer performance across a broader range of regions and conditions. This study lays the groundwork for further investigating sustainable and accurate airport traffic prediction models.

Item Type:	Artikel Umum
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisi / Prodi:	Faculty of Industrial Technology (Fakultas Teknologi Industri) > S1-Electrical Engineering (S1-Teknik Elektro)
Depositing User:	M.Eng. Alfian Ma'arif
Date Deposited:	06 Jan 2025 05:59
Last Modified:	06 Jan 2025 05:59
URI:	http://eprints.uad.ac.id/id/eprint/78129

Actions (login required)

View Item