Prediksi Kemunculan Titik Panas Di Lahan Gambut Provinsi Riau Menggunakan Long Short Term Memory

Ulfa Khaira; Muksin Alfalah; Pikir Claudia Septiani Gulo; Robi Purnomo

doi:10.30591/jpit.v5i3.1931

Prediksi Kemunculan Titik Panas Di Lahan Gambut Provinsi Riau Menggunakan Long Short Term Memory

Ulfa Khaira, Muksin Alfalah, Pikir Claudia Septiani Gulo, Robi Purnomo

Abstract

Indonesia is blessed with the largest and most diverse tropical forests in the world. Millions of Indonesians depend on these forests for their lives. But lately forest fires have become an international concern as an environmental and economic issue. One of the causes of the decline in the number of forests is forest fires. Forest fires produce high particle emissions which can endanger human health. For this reason, necessary precautions. One prevention that can be done is to predict the emergence of hotspots, especially in areas where forest fires are frequent. One way to reduce forest fires is to predict the emergence of hot spots on peatlands with the Long Short Term Memory (LSTM) method. This study predicts the emergence of hotspots in Riau Province over the next 6 months, from August 2019 to January 2020. LSTM is able to predict time series with RMSE 363.38.

Keywords

Hotspot, LSTM, Prediction, Peatland

Full Text:

References

B. Locatelli, M. Brockhaus, M. Colfer, CJP Murdiyarso and H. Santoso., Menghadapi masa depan yang tak pasti: bagaimana hutan dan manusia beradaptasi terhadap perubahan iklim. Bogor: Center for International Forestry Research (CIFOR), 2009.

U. Khaira, I. S. I. S. Sitanggang, and L. Syaufina, “Detection and Prediction of Peatland Cover Changes Using Support Vector Machine and Markov Chain Model,” TELKOMNIKA (Telecommunication Comput. Electron. Control., vol. 14, no. 1, p. 294, 2016.

W. C. Adinugroho, I. N. N. Suryadiputra, B. H. Saharjo, and L. Siboro, Pengendalian Kebakaran Hutan. Bogor: Wetlands International- Indonesia Programme dan Wildlife Habitat Canada, 2005.

R. Sahputra, S. Sutikno, and A. Sandhyavitri, “Mitigasi Bencana Kebakaran Lahan Gambut Berdasarkan Metode Network Analysis Berbasis GIS (Studi Kasus: Pulau Bengkalis),” J. Online Mhs. Fak. Tek. Univ. Riau, vol. 4, no. 2, pp. 1–11, 2017.

I. S. Sitanggang, S. Kirono, and L. Syaufina, “Temporal Patterns of Hotspot Sequences for Early Detection of Peatland Fire in Riau Province,” in Proceedings - 2018 IEEE Asia-Pacific Conference on Geoscience, Electronics and Remote Sensing Technology: Best Practice for Disaster Mitigation using Geoscience, Electronic, and Remote Sensing, AGERS 2018, 2018.

J. Y. Kumar and B. S. Kumar, “Min max normalization based data perturbation method for privacy protection,” Int. J. Comput. Commun. Technol., vol. 2, no. 8, pp. 45–50, 2011.

J. Schmidhuber, “Deep learning in neural networks: An overview,” Neural Networks, vol. 61, pp. 85–117, Jan. 2015.

C. Olah, “Understanding LSTM Networks.” Internet: https://colah.github.io/posts/2015-08-Understanding-LSTMs/, 2015.

D. P. Kingma and J. L. Ba, “Adam: A method for stochastic optimization,” in 3rd International Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings, 2015.

Vitaly Bushaev, “Understanding RMSprop - Faster Neural Network Learning.” Internet: https://towardsdatascience.com/understanding-rmsprop-faster-neural-network-learning-62e116fcf29a, Sept. 12, 2018 [ Nov. 29, 2019]

DOI: https://doi.org/10.30591/jpit.v5i3.1931