在無數的自然災害中，地震能夠在轉瞬間給予人類文明造成巨大的破壞，對地震的認知不足及疏於防範將導致災難及無法抹滅的痛苦和陰影，因此，地震預測在研究主題中一直是重要焦點。而台灣位於環太平洋地震帶上，地震活動頻繁，如何預防地震成為重要的課題。本研究中嘗試對地震預測領域進行概述，並由人工智慧中機器學習領域的理論出發，以人工神經網路的深度學習模型及隨機森林模型，對中央氣象局提供桃園地區的歷史性地震資料進行訓練、預測及分析。本研究包含中央氣象局公開下載之災害性地震資料及購買之資料兩組預測組別，並且後者包含前者，分別以兩種模型預測加速度及震度；此外，本研究也探討將地震資料取對數或平方對預測結果的影響。

Among countless natural disasters, earthquakes can cause huge damage to human civilization in an instant. Insufficient understanding of earthquakes and neglect of prevention will lead to disasters and indelible pain and shadows. Therefore, earthquake prediction has always been an important research topic. Taiwan is located in the Pacific Rim seismic belt, with frequent seismic activity. Prevention of earthquakes disasters has become an important issue. This study will give an overview of the field of earthquake prediction, using machine learning methods. Here, the deep learning model of artificial neural network and random forest model were used to analyzes the historical earthquake data in Taoyuan area provided by Central Weather Bureau. This study considered two prediction data groups, the disastrous earthquake data publicly downloaded by the Central Weather Bureau and the purchased data, the latter including the former, and two models are used to predict acceleration and intensity. In addition, this study also explores the effect of taking the logarithm or square of the earthquake data on the prediction results.

[1]Allen, R. M., Kanamori, H., 2003, “The potential for earthquake early warning in southern California,” Science, Vol. 300, pp. 786-789.
[2]Zollo, A., M. Lancieri, S. Nielsen, 2006, “Earthquake magnitude estimation from peak amplitudes of very early seismic signals on strong motion records,” Geophys. Res. Lett., 33, L23312.
[3]Allen, R. M., Gasparini, P., Kamigaichi, O., and Bose, M., 2009, “The Status of Earthquake Early Warning around the World: An Introductory Overview,” Seismological Research Letters , 80(5), pp. 682–693.
[4]Satriano, C., Wu, Y.-M., Zollo, A., and Kanamori, H, 2011, "Earthquake early warning: Concepts, methods and physical grounds." Soil Dynamics and Earthquake Engineering 31.2, pp. 106-118.
[5]Kohler, M.D., Cochran, E.S., Given, D., Guiwits, S., Neuhauser, D., Henson, I., Hartog, R., Bodin, P., Kress, V., Thompson, S. and Felizardo, C., 2017, ”Earthquake early warning ShakeAlert system: West coast wide production prototype.” Seismological Research Letters 89.1, pp. 99-107.
[6]Minson, S. E., Meier, M. A., Baltay, A. S., Hanks, T. C., and Cochran, E. S., 2018, ”The limits of earthquake early warning: Timeliness of ground motion estimates.” Science advances 4.3 (2018): eaaq0504.
[7]Allen, R.M., Kanamori, H., 2003, “The potential for earthquake early warning in southern California,” Science New Series, Vol. 300, No. 5620.
[8]Japan Meteorological Agency, 2007, “What is the Earthquake Early Warning.” https://www.jma.go.jp/jma/en/Activities/eew1.html
[9]沈哲平, 林主潔, 黃謝恭, 林沛暘, 王冠又, 2010, “類神經網路應用於強震即時警報系統之建物受震反應分析”, 第十屆中華民國結構工程研討會.
[10]Hsu, T.Y., Wu, R.T., Liang, C.W., Kuo, C.H., Lin, C.M., Chang, T.M., Wen, K.L., Loh, C.H., 2013, ” Rapid on-site peak ground acceleration estimation based on support vector regression and P-wave features in Taiwan,” Soil Dynamics and Earthquake Engineering 49, pp. 210-217.
[11]Wu, B.R., Hsiao, N.C., Lin, P.Y., Hsu, T.Y., Chen, C.Y., Hung, S.K., Chiang, H.W., 2017, “An integrated earthquake early warning system and its performance at schools in Taiwan,” Journal of Seismology volume 21, pp. 165–180.
[12]Hsu, T.Y., Wu, R.T., Liang, C.W., Kuo, C.H., Lin, C.M., 2020, ” Peak ground acceleration estimation using P-wave parameters and horizontal-to-vertical spectral ratios,” Terr. Atmos. Ocean. Sci., Vol. 31, No. 1, pp. 1-8.
[13]Reid, H.F., 1906, “The Mechanics of the Earthquake,” The California Earthquake of April 18, 1906, Report of the State Investigation Commission, Vol.2, Carnegie Institution of Washington, Washington, D.C. 1910.
[14]王乾盈 (編), 2014, 基礎地球科學上, 新北市全華出版社, pp. 80.
[15]交通部中央氣象局, 2019, “震度新分級,” 交通部中央氣象局新聞稿.
[16]吳逸民, 蕭乃祺, 林孝維, 2016, “地震初達波強震即時警報系統之研發,” 地震技術報告彙編(交通部中央氣象局). https://web.archive.org/web/20161003005842/http://scweb.cwb.gov.tw/research/51vol/MOTC-CWB-97-E-10.pdf
[17]蕭乃祺, 2015, “中央氣象局強震即時警報作業推動現況,” 台灣地震科學中心通訊. https://tec.earth.sinica.edu.tw/upload/publications/TEC-Newsletter_201506.pdf
[18]交通部中央氣象局, “地震測報APP”. https://www.cwb.gov.tw/V8/C/S/eservice/app/app_e.html#qrcode
[19]交通部中央氣象局, 2016, 105地震年報. https://www.cwb.gov.tw/Data/service/notice/download/publish_20170922120159.pdf
[20]中央研究院, 2015, “大規模地震災害防治策略建議書,” 中央研究院報告No.13, pp. 19-21.
[21]強震即時警報應用交流平台, 2015, “基本原理介紹,” 國家災害防救科技中心. https://web.archive.org/web/20150925063751/http://eew.ncdr.nat.gov.tw/Development_background02.html
[22]Paluszek, M., Thomas, S., 2017, MATLAB Machine Learning, Apress, US.
[23]J. Grus., 2015, Data Science from Scratch. O’Reilly.
[24]Hardesty, L., 2017, “Explained: Neural networks,” MIT News Office. https://news.mit.edu/2017/explained-neural-networks-deep-learning-0414
[25]Wu, D., Jennings, C., Terpenny, J., Gao, R.X., Kumara, S., 2017, “A Comparative Study on Machine Learning Algorithms for Smart Manufacturing: Tool Wear Prediction Using Random Forests,” Journal of Manufacturing Science and Engineering, pp. 3-5.
[26]Lau, S., 2017, “A Walkthrough of Convolutional Neural Network – Hyperparameter Tuning,” Towards Data Science. https://towardsdatascience.com/a-walkthrough-of-convolutional-neural-network-7f474f91d7bd
[27]Bengio, Y., 2009, “Learning Deep Architectures for AI,” Foundations and Trends in Machine Learning. 2 (1), pp. 1–127. https://www.nowpublishers.com/article/Details/MAL-006
[28]Moolayil, J., 2018, Learn Keras for Deep Neural Networks, Apress, US, pp. 4.
[29]Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R., 2014, “Dropout: A Simple Way to Prevent Neural Networks from Overfitting,” Journal of Machine Learning Research 15, pp. 1930-1931.
[30]Myles, A.J., Feudale, R.N., Liu, Y., Woody, N.A., Brown, S.D., 2004, “An introduction to decision tree modeling,” JOURNAL OF CHEMOMETRICS, pp. 275–285.
[31]Aznar, P., 2020, “Decision Trees: Gini vs Entropy,” ETS Asset Management Factory. https://quantdare.com/decision-trees-gini-vs-entropy/
[32]10程式中, 2021, “[Day 12] 決策樹 (Decision tree),” iT邦幫忙技術文章. https://ithelp.ithome.com.tw/articles/10271143
[33]Breiman, L., 2001, “Random Forests,” Mach. Learn., 45(1), pp. 5–32.
[34]Liaw, A., and Wiener, M., 2002, “Classification and Regression by Random Forest,” R News, 2(3), pp. 18–22.
[35]10程式中, 2021, “[Day 14] 多棵決策樹更厲害：隨機森林 (Random forest),” iT邦幫忙技術文章. https://ithelp.ithome.com.tw/articles/10272586
[36]Mao, W., He, J., Tang, J., & Li, Y., 2018, “Predicting remaining useful life of rolling bearings based on deep feature representation and long short-term memory neural network,” Advances in Mechanical Engineering, pp. 1–18.
[37]Zhang, W.J., Yang, G., Lin, Y., Ji, C., & Gupta, M.M., 2018, “On definition of deep learning,” In: Proceedings of the 2018 World Automation Congress (WAC), pp. 1–5.
[38]Kusuma, A.I., Huang, Y.M., 2022, “Product quality prediction in pulsed laser cutting of silicon steel sheet using vibration signals and deep neural network,” Journal of Intelligent Manufacturing, pp. 5.
[39]Baek, J., & Choi,Y., 2019, “Deep neural network for ore production and crusher utilization prediction of truck haulage system in underground mine,” Applied Sciences, 9(19), 4180.
[40]Zhang, R., Peng, Z.,Wu, L., Yao, B., & Guan, Y., 2017, “Fault diagnosis from raw sensor data using deep neural networks considering temporal coherence,” Sensors, 17(3), 549.
[41]Moolayil, J., 2019, Learn keras for deep neural networks: a fast-track approach to modern deep learning with python, pp. 35–38.
[42]Kingma, D.P., & Ba, J., 2015, “Adam: a method for stochastic optimization,” CoRR, abs/1412.6980. https://arxiv.org/abs/1412.6980
[43]Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I. & Salakhutdinov, R., 2014, “Dropout: A simple way to prevent neural networks from overfitting.” Journal of Machine Learning Research, 15, 1929–1958. https://jmlr.org/papers/v15/srivastava14a.html
[44]Zhao, Y., Sun, J., Gupta, M. M., Moody, W., Laverty, W. H., & Zhang, W., 2017, “Developing a mapping from affective words to design parameters for affective design of apparel products,” Textile Research Journal, 87(18), 2224–2232. https://doi.org/10.1177/0040517516669072
[45]Feng, C.-X.J.,Yu, Z.-G.S., Emanuel, J. T., Li, P.-G., et al, 2008, “ Threefold versus fivefold cross-validation and individual versus average data in predictive regression modelling of machining experimental data,” International Journal of Computer Integrated Manufacturing, 21(6), 702–714. https://doi.org/10.1080/09511920701530943
[46]Lin, W.-J., Lo, S.-H., Young, H.-T., & Hung, C.-L., 2019, “Evaluation of deep learning neural networks for surface roughness prediction using vibration signal analysis,” Applied Sciences, 9(7), 1462. https://doi.org/10.3390/app9071462
[47]Wu, T.-Y., & Lei, K. W., 2019, “Prediction of surface roughness in milling process using vibration signal analysis and artificial neural network,” The International Journal of Advanced Manufacturing Technology, 102, 305–314. https://doi.org/10.1007/s00170-018-3176-2
[48]N. J. D. Nagelkerke, 1991, “A Note on a General Definition of the Coefficient of Determination,” Biometrika, Vol. 78, No. 3. (Sep., 1991), pp. 691-692. https://www.jstor.org/stable/2337038?seq=1
[49]魏夢麗,呂秀英, 1999, “決定係數(R^2)在回歸分析中的解釋級正確使用,” 科學農業 47(11,12), pp. 341-345. http://ilc.hk.edu.tw/c/document_library/get_file?p_l_id=260741&folderId=261080&name=DLFE-3350.pdf