PREDIKSI PARAMETER CUACA MENGGUNAKAN DEEP LEARNING LONG-SHORT TERM MEMORY (LSTM)
DOI:
https://doi.org/10.31172/jmg.v21i2.619Keywords:
prediksi, parameter cuaca, deep learning, LSTMAbstract
Saat ini metode deep learning dapat diaplikasikan untuk memprediksi suatu kejadian, seperti memprediksi cuaca suatu wilayah. Salah satu contoh deep learning yang cocok digunakan pada jenis data time series adalah LSTM. Penelitian ini menerapkan metode deep learning LSTM dengan jumlah layer 200, perbandingan data training dengan data test sebesar 9:1, serta mengukur nilai RMSE dan RMSE update hasil validasi dan prediksi beberapa hari ke depan. Data yang digunakan terdiri dari pengukuran suhu udara, kelembaban udara, kecepatan angin, dan tekanan udara selama bulan Januari dan Februari 2019. Data bulan Januari digunakan sebagai data training dan test untuk melakukan validasi prakiraan, sedangkan data bulan Februari digunakan sebagai pembanding dari hasil prediksi deep learning LSTM. Hasil penelitian menunjukkan RMSE seluruh validasi parameter cuaca nilainya semakin baik ketika menggunakan LSTM dengan update. Diperoleh RMSE update untuk parameter suhu, kelembaban, kecepatan angin, dan tekanan udara masing-masing bernilai 0,576; 2,8687; 2,1963; dan 1,0647. Sedangkan prediksi suhu udara, kelembaban, kecepatan angin, dan tekanan udara untuk 1 hari ke depan (1 Februari 2019) masing-masing sebesar 1,0337; 6,3413; 2,8934; dan 1,4313. Dari parameter tersebut hanya parameter suhu dan kelembaban udara yang mengalami pertambahan RMSE seiring bertambahnya waktu. Sedangkan parameter kecepatan angin dan tekanan udara mengalami penurunan di hari ketiga dan meningkat secara kontinu hingga satu bulan ke depan.
Today, deep learning can be applied to predict any events, such as predict the weather of a region. One of them is LSTM which is suitable for use in time series data types. This study conducted the deep learning LSTM with the number of 200 layers, ratio training with test data of 9:1, measuring the value of validation RMSE and RMSE update and also predictions some weather parameters in a few days later. The data used consisted of measurements of air temperature, humidity, wind speed, and air pressure during January and February 2019. The data January were used as training and test data to conduct forecast validation, while the data February was used as a comparison of the results predicted for deep learning LSTM. The result shows that the forecast RMSE for all-weather parameters is better when using LSTM with an update. Obtained for temperature, relative humidity, wind speed, and air pressure have RMSE with update are 0,576; 2,8687; 2,1963; and 1,0647, respectively. While the prediction of air temperature, wind speed, and air pressure for one day later (1 February 2019) is 1.0337; 6.3413; 2,8934; and 1.4313, respectively. From the all-weather parameters only temperature and humidity parameter that increase in RMSE over time. While the parameters such as wind speed and air pressure decreased on the third day and increased continuously for the next one month.
References
BMKG. Laporan Indeks Kepuasan Masyarakat 2018, Pusat Meteorologi Publik BMKG.
Kemenkeu. Daftar Isian Pelaksanaan Anggaran (DIPA) Induk BMKG Tahun Anggaran 2018.
G. Hinton, et al, "Deep Neural Networks for Acoustic Modeling in Speech Recognition," IEEE Signal Processing Magazine, pp.82-97, 2012.
G.E. Hinton, S. Osindero, and Y-W Teh, "A Fast Learning Algorithm for Deep Belief Nets," Journal Neural Computation, vol. 18(7), 1527 - 1554, 2006.
K. Simonyan and A. Zisserman, ”Very Deep Convolutional Networks for Large-Scale Image Recognition," ICLR, 2015.
K. He, X. Zhang, S. en, and J. Sun, "Deep Residual Learning for Image Recognition," IEEE Conference on Computer Vision and Pattern Recognition, 2015.
Y, Bengio, A. Courville, and P. Vincent, "Representation Learning: A Review and New Perspectives," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.35 (8), pp. 1798–1828, 2013.
J. Schmidhuber, "Deep Learning in Neural Networks: an Overview," Neural Networks, vol. 61 pp. 85–117, 2015.
Y. Bengio, Y. LeCun, G. Hinton, "Deep Learning," Nature, vol. 521 (7553), pp. 436–444, 2015.
G, Chen, "A Gentle Tutorial of Recurrent Neural Network with Error Backpropagation," Internet: https://arxiv.org/pdf/1610.02583, diakses 16 Maret 2019.
H, Sak, A, Senior, and F, Beaufays, "Long Short-Term Memory Recurrent Neural Network Architectures for Large Scale Acoustic Modeling", Interspeech, 2014.
The Clever Machine, "Derivation: Derivatives for Common Neural Network Activation Functions," Internet: https://theclevermachine.wordpress.com/2014/09/08/derivation-derivatives-for-common-neural-network-activation-functions/, diakses 22 Maret 2019.
Nvidia, "Long-Short Term Memory (LSTM)". Internet https://developer.nvidia.com/discover/lstm, diakses 16 Maret 2019.
S. Hochreiter and J. Schmidhuber, "Long Short-Term Memory," Neural Computation, vol. 9(8), pp. 1735-1780, 1997.
S.A. Haider et al, "LSTM Neural Network Based Forecasting Model for Wheat Production in Pakistan," Agronomy, vol. 9(2), 2019.
D. S. Reddy and P. R. C. Prasad, "Prediction of Vegetation Dynamics Using NDVI Time Series Data and LSTM," Modelling earth systems and Environment, vol. 4(1), pp. 409 - 419, 2018.
J. Zhang, Y. Zhu, X. Zhanf, M. Ye, and J. Yang, "Developing a Long Short-Term Memory (LSTM) Based Model for Predicting Water Table in Agricultural Areas," Journal of Hydrology, vol. 561, pp. 918 -929, 2018.
Z. Jiang, et al, "Predicting County Level Corn Yields Using Deep Long Short Term Memory Models," Internet: https://arxiv.org/abs/1805.12044, diakses 17 Maret 2018.
J. Shook, L. Wu, T. Gangopadhyay, B. Ganapathysubramanian, and S. Sarkar, "Integrating Genotype And Weather Variables for Soybean Yield Prediction Using Deep Learning," Iowa State University, 2018.
X. Li, "Pollutant Concentration Predictions: Method Development And Evaluation," Environmental Pollution, vol. 231, pp. 997-1004, 2017.
C-J. Huang and P.H.Kuo, " A Deep CNN-LSTM Model for Particulate Matter (PM2.5) Forecasting in Smart Cities", Sensors vol.18(7), 2018.
M.Sundermeyer, R. Schluter, and H. Ney, "LSTM Neural Networks for Language Modelling," Interspeech 2012, 2012.
M.Sudermeyer, H. Ney, and R. Schluter," From Feedforward to Recurrent LSTM Neural Networks for Language Modeling," IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 23(3), 2015.
O. Vinyals, et al," Grammar as Foreign Language," Advances in Neural Information Processing Systems 28 (NIPS 2015), 2015.
Y, Kim, Y. Jernite, D. Sontag, and A. M. Rush, "Character-Aware Neural Language Models," Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence (AAAI-16), 2016.
H. Bakir, G. Chniti, and H. Zaher, "E-Commerce Price Forecasting Using LSTM Neural Networks," International Journal of Machine Learning and Computing, vol.18 (2), pp. 169 - 174, 2018.
G. Chniti, H. Bakir, and H. Zaher, " E-commerce Time Series Forecasting using LSTM Neural Network and Support Vector Regression," Proceedings of the International Conference on Big Data and Internet of Thing, pp. 80 - 84, 2017.
X. Shi, Z. Chen, H. Wang, and D-Y. Yeung, "Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting," Internet: https://arxiv.org/abs/1506.04214, diakses 17 Maret 2019.
M.A Zaytar and C. E. Amrani, " Sequence to Sequence Weather Forecasting with Long Short-Term Memory Recurrent Neural Networks," International Journal of Computer Applications, vol. 143(11), pp. 7-11, 2016.
L, Wiest, ”Recurrent Neural Networks Combination of RNN and CNN,” Internet: https://wiki.tum.de/display/lfdv/Recurrent+Neural+Networks+-+Combination+of+RNN+and+CNN, diakses 20 April 2019.
Z. Reitermanova, "Data Splitting," WDS'S Proceedings of Contributed Papers, part I, pp. 31-36, 2010.
A. Dey, "Machine Learning Algorithms: A Review," International Journal of Computer Science and Information Technologies", vol. 7(3), pp. 1174-1179, 2016.
I.I. Tika, "Variasi Suhu dan Kelembaban Udara di Taman Suropati dan Sekitarnya," Skripsi, Fakultas MIPA: Universitas Indonesia, 2010.
R. Anggreni, Muliadi, and R. Adriat, "Analisis Pengaruh Tutupan Awsn terhadap Radiasi Matahari di Kota Pontianak," Prisma Fisika, vol. 3, pp.214-219, 2018.
G.E.P. Box and R.D. Meyer, "An Analysis for Unreplicated Fractional Factorials," Technometrics, vol. 28(1), 1986.
M.E.J Newman, "Power Laws, Pareto Distributions and Zipf's Law," Internet: https://arxiv.org/PS_cache/cond-mat/pdf/0412/0412004v3.pdf, diakses 21 Maret 2019.
V. Gholami, K.W. Chau, F. Fadaee, J. Tokaman, and A. Ghaffari, "Modeling Of Groundwater Level Fluctuations Using Dendrochronology in Alluvia Aquifers," Journal of Hydrology, vol. 529 (3), pp. 1060-1069, 2015.
N.P.I. Wikandari, H. Harsa, A.F. Assomadi ,dan R. Boedisantoso, "Penentuan Korelasi Perubahan Kecepatan Angin dan Kekuatan Radiasi terhadap Ketinggian Lapisan inversi dan Hubungannya dengn Kualitas Udara Ambien Kota Surabaya," Jurnal Teknik ITS, vol. 4(1), pp. F120-F124, 2015.
N.S. Keskar, J. Nocedal, P.T.P. Tang, D. Mudigere, and M. Smelyanskiy, "On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima," Internet: https://arxiv.org/abs/1609.04836, diakses 24 Maret 2019.
A. Tato, R. Nkambou, and C. Frasson, "Predicting from Multimodal User Data using a Deep Long Term Memory," UMAP, 2018, Internet: https://www.researchgate.net/publication/326212084_Predicting_Emotions_From_Multimodal_Users'_Data, diakses 24 Maret 2019.
Y. Hua, et al, "Deep Learning with Long Short-Term Memory for Time Series Prediction," Internet: https://arxiv.org/pdf/1810.10161, diakses 25 Maret 2019.
E. Hoffer, I. Hubara, and D. Soudry, "Train Longer, generalize better: Closing the Generalization Gap in Large Batch Training of Neural Networks," Internet: https://arxiv.org/abs/1705.08741, diakses 25 Maret 2019.
S.E. Reed et al, ”Training Deep Neural Network on noisy Labels with Bootstrapping," Internet: https://arxiv.org/abs/1412.6596, diakses 24 Maret 2019.
Kirtman, et al, "Near-term Climate Change:Projections and Predictability," Cambridge University Press.
Zhang, et al, "What Is the Predictability Limit of Midlatitude Weather," Journal of the Atmospheric Sciences, vol. 76, pp. 1077 - 1091, 2019.
Zhu, et al., "Seamless Precipitation Prediction Skill in the Tropics and Extratropics from a Global Model," Monthly Weather Review, vol. 142, pp. 1556 - 1569, 2014.
Downloads
Additional Files
Published
How to Cite
Issue
Section
License
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
