Evaluasi Mekanisme Attention pada Arsitektur Bidirectional LSTM dan GRU untuk Prediksi Intensitas dan Trajektori Siklon Tropis
Abstract
Penelitian ini mengevaluasi kontribusi mekanisme attention dalam meningkatkan akurasi prediksi intensitas dan trajektori siklon tropis di Pasifik Barat menggunakan tiga konfigurasi hyperparameter dari literatur, yaitu Nketiah (2023), Giffard-Roisin (2020), dan Mulia (2025). Enam arsitektur Recurrent Neural Network (RNN) yaitu LSTM, BiLSTM, Att-BiLSTM, GRU, BiGRU, dan Att-BiGRU. Model ini dilatih menggunakan data IBTrACS beresolusi 3 jam periode 1945-2025. Kinerja model dievaluasi dengan MAE, RMSE, dan MAPE pada lead time 12, 24, 48, dan 72 jam. Hasil menunjukkan bahwa model berbasis attention secara konsisten menghasilkan error lebih rendah dibandingkan model tanpa attention pada prediksi intensitas. Konfigurasi Nketiah (2023) dengan Att-BiLSTM memberikan performa terbaik dengan MAE intensitas 7,78 knot dan MAE trajektori 77,97 km pada lead time 12 jam. Konfigurasi Giffard-Roisin (2020) dengan Att-BiGRU unggul pada prediksi intensitas jangka pendek hingga menengah, sedangkan konfigurasi Mulia (2025) menghasilkan error intensitas terendah namun akurasi posisi lebih rendah. Analisis kasus menunjukkan attention efektif pada siklon dengan pergerakan gradual, tetapi masih terbatas pada siklon dengan translasi zonal cepat. Temuan ini mendukung penerapan attention dalam sistem peringatan dini siklon tropis operasional.
Downloads
References
D. Bahdanau, K. Cho, and Y. Bengio, “Neural machine translation by jointly learning to align and translate,” arXiv preprint arXiv:1409.0473, 2016.
J. Brownlee, Long Short-Term Memory Networks With Python. Melbourne, Australia: Machine Learning Mastery, 2017.
K. Cho, B. van Merrienboer, D. Bahdanau, and Y. Bengio, “On the properties of neural machine translation: Encoder-decoder approaches,” arXiv preprint arXiv:1409.1259, 2014.
J. Chung, C. Gulcehre, K. Cho, and Y. Bengio, “Empirical evaluation of gated recurrent neural networks on sequence modeling,” arXiv preprint arXiv:1412.3555, 2014.
R. L. Elsberry, “Advances in research and forecasting of tropical cyclones from 1963–2013,” Asia-Pacific Journal of Atmospheric Sciences, vol. 50, no. 1, pp. 3–16, 2014, doi: 10.1007/s13143-014-0013-y.
K. Emanuel, “100 years of progress in tropical cyclone research,” Meteorological Monographs, vol. 59, pp. 15.1–15.68, 2018, doi: 10.1175/AMSMONOGRAPHS-D-18-0016.1.
S. Giffard-Roisin, M. Yang, G. Charpiat, C. Kumler Bonfanti, B. Kégl, and C. Monteleoni, “Tropical cyclone track forecasting using fused deep learning from aligned reanalysis data,” Frontiers in Big Data, vol. 3, 2020, doi: 10.3389/fdata.2020.00001.
S. M. Griffin, A. Wimmers, and C. S. Velden, “Predicting rapid intensification in North Atlantic and Eastern North Pacific tropical cyclones using a convolutional neural network,” Weather and Forecasting, vol. 37, no. 8, pp. 1333–1355, 2022, doi: 10.1175/WAF-D-21-0201.1.
S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997, doi: 10.1162/neco.1997.9.8.1735.
T. O. Hodson, “Root-mean-square error (RMSE) or mean absolute error (MAE): when to use them or not,” Geoscientific Model Development, vol. 15, no. 14, pp. 5481–5487, 2022, doi: 10.5194/gmd-15-5481-2022.
H. Huang, D. Deng, L. Hu, Y. Chen, and N. Sun, “Beyond the backbone: A quantitative review of deep-learning architectures for tropical cyclone track forecasting,” Remote Sensing, vol. 17, no. 15, 2025, doi: 10.3390/rs17152620.
A. Tholib, N. K. Agusmawati, and F. Khoiriyah, “Prediksi Harga Emas Menggunakan Metode LSTM dan GRU,” Jurnal Informatika dan Teknik Elektro Terapan, vol. 11, no. 3, pp. 620–627, Aug. 2023, doi: 10.23960/jitet.v11i3.3250.
T. Knutson, J. Camargo, J. C. L. Chan, K. Emanuel, C.-H. Ho, J. Kossin, K. Walsh, and L. Wu, “Tropical cyclones and climate change assessment Part I: Detection and attribution,” Bulletin of the American Meteorological Society, vol. 100, no. 10, pp. 1987–2007, 2019, doi: 10.1175/BAMS-D-18-0189.1.
J. S. Kumar, V. Venkataraman, S. Meganathan, and K. Krithivasan, “Tropical cyclone intensity and track prediction in the Bay of Bengal using LSTM-CSO method,” IEEE Access, vol. 11, pp. 81613–81622, 2023, doi: 10.1109/ACCESS.2023.3298997.
I. E. Mulia, U. Shimada, N. Ueda, T. Miyoshi, and M. T. Maulana, “Multi-horizon prediction of tropical cyclone intensity and its interpretability with temporal fusion transformer,” Scientific Reports, vol. 15, no. 1, 2025, doi: 10.1038/s41598-025-87000-7.
E. Mulyana, M. B. R. Prayoga, A. Yananto, S. Wirahma, E. Aldrian, B. Harsoyo, T. H. Seto, and Y. Sunarya, “Tropical cyclones characteristic in southern Indonesia and the impact on extreme rainfall event,” MATEC Web of Conferences. Les Ulis, France: EDP Sciences, 2018.
B. S. Naik, V. C. Karthik, B. Manjunatha, G. H. Veershetty, B. S. Varshini, and S. G. Rao, “Stock price forecasting using N-Beats deep learning architecture,” Journal of Scientific Research and Reports, vol. 30, no. 9, pp. 483–494, 2024.
E. A. Nketiah, L. Chenlong, J. Yingchuan, and S. A. Aram, “Recurrent neural network modeling of multivariate time series and its application in temperature forecasting,” PLoS ONE, vol. 18, no. 5, 2023, doi: 10.1371/journal.pone.0281828.
S. Rahman, M. F. Faisal, P. Mondal, and M. M. Rahman, “Tropical cyclone track prediction harnessing deep learning algorithms: A comparative study on the Northern Indian Ocean,” Results in Engineering, vol. 26, 2025, doi: 10.1016/j.rineng.2025.104491.
M. Schuster and K. K. Paliwal, “Bidirectional recurrent neural networks,” IEEE Transactions on Signal Processing, vol. 45, no. 11, pp. 2673–2681, 1997, doi: 10.1109/78.650093.
T. Song, Y. Li, F. Meng, P. Xie, and D. Xu, “A novel deep learning model by BiGRU with attention mechanism for tropical cyclone track prediction in the Northwest Pacific,” Journal of Applied Meteorology and Climatology, vol. 61, no. 1, pp. 3–12, 2022, doi: 10.1175/JAMC-D-21-0020.1.
L. Wang, B. Wan, S. Zhou, H. Sun, and Z. Gao, “Forecasting tropical cyclone tracks in the northwestern Pacific based on a deep-learning model,” Geoscientific Model Development, vol. 16, no. 8, pp. 2167–2179, 2023, doi: 10.5194/gmd-16-2167-2023.
World Meteorological Organization, Glossary. Tropical Cyclone Programme. Geneva, Switzerland: WMO, 2017. [Online]. Available: https://cyclone.wmo.int/pdf/Glossary.pdf
D.-L. Zhang and H. Chen, “Importance of the upper-level warm core in the rapid intensification of a tropical cyclone,” Geophysical Research Letters, vol. 39, no. 2, 2012, doi: 10.1029/2011GL050578.
F. Chollet, Deep Learning with Python, 2nd ed. Shelter Island, NY, USA: Manning Publications, 2021.
Z. C. Lipton, J. Berkowitz, and C. Elkan, “A critical review of recurrent neural networks for sequence learning,” arXiv preprint arXiv:1506.00019, 2015.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
