中國武漢自2019年12月爆出新型冠狀病毒疫情，隨後迅速擴散至中國其他省分及世界各地，而臺灣在2020年1月21日傳出首例境外移入確診病例後，每天都有出現確診案例，截至7月31日為止，臺灣新型冠狀病毒之累積確診人數為467例，其中376例為境外移入，高達80%。
本研究蒐集COVID-19累積通報人數、累積復原人數、臺灣機場累積入境人數、美國入境臺灣班機數與歐洲入境臺灣班機數作為自變數，以長短期記憶與差分整合移動平均自迴歸模型作為預測分析模型，對臺灣COVID-19境外移入病例數進行多變量分析，並利用過去的歷史資料來預測未來的走勢。
研究結果發現以三天作為滑動窗格大小來預測臺灣COVID-19境外移入確診人數最為準確，且累積通報人數、美國入境台灣班機數及歐洲入境臺灣班機數這三個變數可做為預測臺灣COVID-19的境外移入確診病例數之主要自變數。而長短期記憶模型與差分整合移動平均自迴歸模型相比，在預測未來臺灣COVID-19的趨勢上可以得到較低的誤差值。
本研究使用深度學習對臺灣新冠肺炎境外移入的確診人數進行預測分析，提供未來臺灣新冠肺炎研究一個可行的方向，政府相關單位可依此作為參考，盡速控制新型冠狀肺炎在臺灣的傳播狀況。

Since the outbreak of the new coronavirus in Wuhan, China in December 2019, it quickly spread to other provinces in China and around the world. After the first imported case of 2019 novel coronavirus infection in Taiwan on January 21, 2020, confirmed cases appeared every day. As of July 31, 2020, the cumulative number of confirmed cases of 2019 novel coronavirus infection (COVID-19) in Taiwan was 467, of which 376 are imported cases, up to 80%.
We collected the cumulative number of COVID-19 reported cases in Taiwan, recovered cases in Taiwan, arrival passengers in airports in Taiwan, flights from Europe to Taiwan, and flights from the USA to Taiwan as independent variables. Using Long short-term memory and Autoregressive Integrated Moving Average model as predicting models to conduct the multivariate analysis of the number of COVID-19 imported cases in Taiwan, and using historical data to forecast future trends. The results suggest that 3-day is the most accurate sliding window size to forecast the number of COVID-19 imported cases in Taiwan, and the cumulative number of COVID-19 reported cases in Taiwan, flights from Europe to Taiwan and flights from the USA to Taiwan are the main independent variables. Besides, using Long short-term memory model to forecast COVID-19 trends in Taiwan can get a lower error than Autoregressive Integrated Moving Average model.
We use deep learning to forecast the number of COVID-19 imported cases in Taiwan and provide a feasible direction for future research of COVID-19 in Taiwan. The government agencies can use this as a reference to control the spread of COVID-19 in Taiwan as soon as possible.

論文審定書 i
摘要 ii
Abstract iii
第一章、緒論 1
第一節、研究背景與動機 1
第二節、研究目的 6
第三節、研究流程 6
第二章、文獻探討 8
第一節、機器學習應用於預測傳染病 8
第二節、深度學習應用於預測傳染病 10
第三節、長短期記憶應用於預測傳染病 11
第四節、COVID-19的相關研究 15
一、 易感染－暴露－感染者－恢復模型 15
二、 整合移動平均自回歸模型 18
三、 長短期記憶 19
四、 COVID-19相關研究之變數整理 21
第三章、研究方法 22
第一節、變數選擇與說明 22
第二節、資料來源 23
第三節、資料前處理 24
一、 數據正規化 24
二、 變數組合 25
三、 滑動窗格法 26
第四節、資料分析 28
一、 LSTM模型 28
二、 ARIMA模型 30
三、 平均絕對百分比誤差（MAPE） 31
第四章、研究結果 32
第一節、模型擬合 32
一、 LSTM模型擬合 32
二、 LSTM模型擬合結果整理 42
三、 ARIMA模型擬合 43
第二節、模型擬合結果比較 48
第三節、模型預測結果 49
一、LSTM模型預測 49
二、ARIMA模型預測 50
三、模型預測結果比較 51
第五章、結論與建議 52
第一節、研究發現 52
第二節、研究貢獻 53
第三節、研究限制與未來建議 55
參考文獻 56
中文文獻 56
英文文獻 59

中文文獻
BBC (2020)。武漢肺炎：美國西雅圖確診第一例。2020年7月1日，取自：https://www.bbc.com/zhongwen/trad/world-51199732
人文及社會科學研究發展司 (2018)。動態「時空演算法」—掌握傳染病感染資訊。2020年1月6日，取自：https://www.most.gov.tw/folksonomy/detail?subSite=main&article_uid=d5cb964d-d945-494d-9cd9-a4b0b0f3d769&menu_id=9aa56881-8df0-4eb6-a5a7-32a2f72826ff&l=CH&utm_source=rss
三立新聞網 (2020)。武漢肺炎擴散！全球出現首宗境外病例　華婦泰國確診感染。2020年6月29日，取自：https://www.setn.com/News.aspx?NewsID=671895
工研院資通所 (2018)。深度學習的訓練資料準備與平台之演進發展。2020年1月6日，取自：https://ictjournal.itri.org.tw/Content/Messagess/contents.aspx?&MmmID=654304432061644411&CatID=654313611255143006&MSID=1001517067307416615
東方新聞網 (2020)。武漢肺炎：韓確診首宗感染個案 患者為中國籍女子。2020年6月29日，取自：https://hk.on.cc/hk/bkn/cnt/aeanews/20200120/bkn-20200120121833575-0120_00912_001.html
林大貴(2017)。Keras深度學習簡介。2020年6月29日，取自：http://tensorflowkeras.blogspot.com/2017/08/keras.html
林大貴(2017)。TensorFlow+Keras深度學習人工智慧實務應用。臺灣：博碩
林東清(2018)。資訊管理: e化企業的核心競爭能力。臺灣：元照出版
移民署內政資料開放平台(2020)。各機場、港口入出境人數統計資料。2020年6月29日，取自：https://data.moi.gov.tw/MoiOD/Data/DataDetail.aspx?oid=905908DA-0EF6-4B24-87B0-35B7EDA4CFD2
楊朝棟、陳源安、林澔、詹毓哲、詹惟臣、劉伯瑜、詹毓偉(2019)。應用長短期記憶模型之深度學習技術於類流感疫情預測。「TANET 2019-臺灣網際網路研討會」發表之論文，高雄國際會議中心。
蔡宜臻(2020)。赴美國參加傳染病疫情監視及預測模式研習訓練。2020年7月6日，取自：https://report.nat.gov.tw/ReportFront/PageSystem/reportFileDownload/C10102095/001
衛生福利部(2019)。新穎智慧辨蚊系統用以監控與追蹤登革熱病媒蚊密度。2020年7月21日，取自：https://www.mohw.gov.tw/fp-4261-46001-1.html
衛生福利部疾病管制署(2018)。疾管署攜手宏碁運用人工智慧推出流感預報站，即時掌握未來疫情動態。2020年7月21日，取自：https://www.cdc.gov.tw/Category/ListContent/AHwuigegBBBmuDcbWkzoGQ?uaid=f9nXXPLjkHuC-lE0V4s8Wg
衛生福利部疾病管制署(2020)。2月10日起經中港澳轉機得入境者，需居家檢疫14天；並限縮我國直航中港澳航線，部分機場暫停航班。2020年8月10日，取自：https://www.cdc.gov.tw/Bulletin/Detail/3TSeMBtW8Gz0vB_UR1RjUA?typeid=9
衛生福利部疾病管制署(2020)。3月24日至4月7日，我國全面禁止旅客登機來台轉機。2020年8月10日，取自：https://www.cdc.gov.tw/Bulletin/Detail/LbQTPXUJx1SBy_DDRSnTMw?typeid=9
衛生福利部疾病管制署(2020)。我國藉由登機檢疫即時發現首例中國大陸武漢移入之嚴重特殊傳染性肺炎個案，指揮中心提升中國大陸武漢之旅遊疫情建議至第三級警告(Warning)。2020年7月1 日，取自：https://www.cdc.gov.tw/Bulletin/Detail/6oHuoqzW9e_onW0AaMEemg?typeid=9
衛生福利部疾病管制署(2020)。新聞稿。2020年6月29日，取自：https://www.cdc.gov.tw/Bulletin/List/MmgtpeidAR5Ooai4-fgHzQ
衛生福利部疾病管制署(2020)。嚴重特殊傳染性肺炎。2020年6月29日，取自：https://www.cdc.gov.tw/Disease/SubIndex/N6XvFa1YP9CXYdB0kNSA9A
衛生福利部疾病管制署(2020)。嚴重特殊傳染性肺炎疾病介紹。2020年6月29日，取自：https://www.cdc.gov.tw/Category/Page/vleOMKqwuEbIMgqaTeXG8A
衛生福利部疾病管制署(2020)。嚴重特殊傳染性肺炎疾病介紹。2020年7月6日，取自：https://www.cdc.gov.tw/Category/Page/vleOMKqwuEbIMgqaTeXG8A
聯合新聞網(2020)。日本出現武漢肺炎首例！患者為中國籍曾前往湖北。2020年6月29日，取自：https://udn.com/news/story/6809/4290661

英文文獻
Acuna-Zegarra, M. A., Comas-Garcia, A., Hernandez-Vargas, E., Santana-Cibrian, M., & Velasco-Hernandez, J. X. (2020). The SARS-CoV-2 epidemic outbreak: a review of plausible scenarios of containment and mitigation for Mexico. medRxiv, 2020.2003.2028.20046276. doi:10.1101/2020.03.28.20046276
Burdakov, A., Ukharov, A., Myalkin, M., & Terekhov, V. (2018). Forecasting of influenza-like illness incidence in amur region with neural networks. Paper presented at the International Conference on Neuroinformatics.
Ceylan, Z. (2020). Estimation of COVID-19 prevalence in Italy, Spain, and France. Science of the total Environment, 138817.
Chen CM, Jyan HW, Chien SC, Jen HH, Hsu CY, Lee PC, Lee CF, Yang YT, Chen MY, Chen LS, Chen HH, Chan CC. (2020). Containing COVID-19 Among 627,386 Persons in Contact With the Diamond Princess Cruise Ship Passengers Who Disembarked in Taiwan: Big Data Analytics. J Med Internet Res 2020;22(5):e19540.
Chimmula, V. K. R., & Zhang, L. (2020). Time series forecasting of COVID-19 transmission in Canada using LSTM networks. Chaos, Solitons & Fractals, 109864.
Chumachenko, D., Turiy, A., & Chukhray, A. (2019). Application of Statistical Simulation for Measles Epidemic Process Forecasting. Paper presented at the 2019 IEEE 2nd Ukraine Conference on Electrical and Computer Engineering (UKRCON).
Copeland, M. (2016). What’s the Difference Between Artificial Intelligence, Machine Learning and Deep Learning? Retrieved from https://blogs.nvidia.com/blog/2016/07/29/whats-difference-artificial-intelligence-machine-learning-deep-learning-ai/
Dziugys, A., Bieliunas, M., Skarbalius, G., Misiulis, E., & Navakas, R. (2020). Simplified model of Covid-19 epidemic prognosis under quarantine and estimation of quarantine effectiveness. medRxiv.
Ferguson, N., Cummings, D., Cauchemez, S. et al. Strategies for containing an emerging influenza pandemic in Southeast Asia. Nature 437, 209–214 (2005). https://doi.org/10.1038/nature04017
Flightradar24 (2020). Flightradar24: Live Flight Tracker. Retrieved from https://www.flightradar24.com/data
Flightstats By Cirium. (2020). Flightstats. Retrieved from https://www.flightstats.com/v2/historical-flight/subscribe
Hethcote, H. W. (2000). The mathematics of infectious diseases. SIAM review, 42(4), 599-653.
Hewamalage, H., Bergmeir, C., & Bandara, K. (2019). Recurrent neural networks for time series forecasting: Current status and future directions. arXiv preprint arXiv:1909.00590.
Hochreiter, S., & Schmidhuber, J. (1997). Long Short-Term Memory. Neural Computation, 9(8), 1735-1780. doi:10.1162/neco.1997.9.8.1735
Hota, H., Handa, R., & Shrivas, A. (2017). Time series data prediction using sliding window based rbf neural network. International Journal of Computational Intelligence Research, 13(5), 1145-1156.
Huang, Wei-Hsuan & Teng, Ling-Chiao & Yeh, Ting-Kuang & Chen, Yu-Jen & Lo, Wei-Jung & Wu, Ming-Ju & Chin, Chun-Shih & Tsan, Yu-Tse & Lin, Tzu-Chieh & Chai, Jyh-Wen & Lin, Chin-Fu & Tseng, Chien-Hao & Liu, Chia-Wei & Wu, Chi-Mei & Chen, Po-Yen & Shi, Zhi-Yuan & Liu, po-yu. (2020). 2019 novel coronavirus disease (COVID-19) in Taiwan: Reports of two cases from Wuhan, China. Journal of Microbiology, Immunology and Infection. 53. 10.1016/j.jmii.2020.02.009.
Jean, S.-S., Lee, P.-I., & Hsueh, P.-R. (2020). Treatment options for COVID-19: The reality and challenges. Journal of Microbiology, Immunology and Infection.
Jia, W., Li, X., Tan, K., & Xie, G. (2019). Predicting the outbreak of the hand-foot-mouth diseases in China using recurrent neural network. In 2019 IEEE International Conference on Healthcare Informatics (ICHI) (pp. 1-4). IEEE.
Jia, W., Wan, Y., Li, Y., Tan, K., Lei, W., Hu, Y., Ma, Z., Li, X., & Xie, G. (2019). Integrating Multiple Data Sources and Learning Models to Predict Infectious Diseases in China. AMIA Joint Summits on Translational Science proceedings. AMIA Joint Summits on Translational Science, 2019, 680–685.
Jiang, D., Hao, M., Ding, F., Fu, J., & Li, M. (2018). Mapping the transmission risk of Zika virus using machine learning models. Acta tropica, 185, 391-399.
Johns Hopkins University. (2020). COVID-19 Data Repository. Retrieved from https://www.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6
Keras. (n.d.). EarlyStopping. Retrieved from https://keras.io/api/callbacks/early_stopping/
Kermack, W. O., & McKendrick, A. G. (1927). A contribution to the mathematical theory of epidemics. Proceedings of the royal society of london. Series A, Containing papers of a mathematical and physical character, 115(772), 700-721.
Kim, S.-Y., Min, K.-D., Lee, S., & Park, S. (2019). Development of a Recurrent Neural Network Model for Prediction of Dengue Importation. Online Journal of Public Health Informatics, 11(1).
Lauer, S. A., Grantz, K. H., Bi, Q., Jones, F. K., Zheng, Q., Meredith, H. R., Azman, A. S., Reich, N. G., & Lessler, J. (2020). The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly Reported Confirmed Cases: Estimation and Application. Annals of internal medicine, 172(9), 577–582. https://doi.org/10.7326/M20-0504
LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436-444.
Lipton, Z. C., Berkowitz, J., & Elkan, C. (2015). A critical review of recurrent neural networks for sequence learning. arXiv preprint arXiv:1506.00019.
Lu, F. S., Hattab, M. W., Clemente, C. L., Biggerstaff, M., & Santillana, M. (2019). Improved state-level influenza nowcasting in the United States leveraging Internet-based data and network approaches. Nature communications, 10(1), 147.
Ng, A., Katanforoosh, K., & Mourri, Y. B. (2018). 深度學習專項課程. Retrieved from https://www.coursera.org/specializations/deep-learning#about
Rangarajan, P., Mody, S. K., & Marathe, M. (2019). Forecasting dengue and influenza incidences using a sparse representation of Google trends, electronic health records, and time series data. PLoS computational biology, 15(11), e1007518.
Reuters. (2020). Wuhan lockdown 'unprecedented', shows commitment to contain virus: WHO representative in China. Retrieved from https://www.reuters.com/article/us-china-health-who-idUSKBN1ZM1G9
Santosh, T., Ramesh, D., & Reddy, D. (2020). LSTM based prediction of malaria abundances using big data. Computers in Biology and Medicine, 103859.
Schäfer, A. M., & Zimmermann, H. G. (2006). Recurrent Neural Networks Are Universal Approximators, Berlin, Heidelberg.
ScienceDaily. (2019). Harnessing multiple data streams and artificial intelligence to better predict flu. Retrieved from https://www.sciencedaily.com/releases/2019/01/190111143744.htm
Shao-Chung Cheng, Yuan-Chia Chang, Yu-Long Fan Chiang, Yu-Chan Chien, Mingte Cheng, Chin-Hua Yang, Chia-Husn Huang, Yuan-Nian Hsu. (2020). First case of Coronavirus Disease 2019 (COVID-19) pneumonia in Taiwan. Journal of the Formosan Medical Association,Volume 119, Issue 3, Pages 747-751.
Sharma, V., Kumar, A., Lakshmi Panat, D., & Karajkhede, G. (2015). Malaria outbreak prediction model using machine learning. International Journal of Advanced Research in Computer Engineering & Technology (IJARCET), 4(12).
Statistical Machine Learning Lab at UCLA. (2020). Learning Epidemic Models for COVID-19. Retrieved from https://covid19.uclaml.org/model.html
Tandon, H., Ranjan, P., Chakraborty, T., & Suhag, V. (2020). Coronavirus (COVID-19): ARIMA based time-series analysis to forecast near future. arXiv preprint arXiv:2004.07859.
Teng, Y., Bi, D., Guo, X., & Paul, R. (2018). Predicting the Epidemic Potential and Global Diffusion of Mosquito-Borne Diseases Using Machine Learning. Available at SSRN 3260785.
Tomar, A., & Gupta, N. (2020). Prediction for the spread of COVID-19 in India and effectiveness of preventive measures. Science of the total Environment, 138762.
Walsh, D. P., Ma, T. F., Ip, H. S., & Zhu, J. (2019). Artificial intelligence and avian influenza: Using machine learning to enhance active surveillance for avian influenza viruses. Transboundary and emerging diseases.
World Health Organization. (2020). WHO Coronavirus Disease (COVID-19) Dashboard. Retrieved from https://covid19.who.int/
Worldometers. (2020). COVID-19 CORONAVIRUS PANDEMIC. Retrieved from https://www.worldometers.info/coronavirus/
Xu, J., Xu, K., Li, Z., Tu, T., Xu, L., & Liu, Q. (2019). Developing a dengue forecast model using Long Short Term Memory neural networks method. bioRxiv, 760702.
Yahmed, Y. B., Bakar, A. A., Hamdan, A. R., Ahmed, A., & Abdullah, S. M. S. (2015). Adaptive sliding window algorithm for weather data segmentation. Journal of Theoretical and Applied Information Technology, 80(2), 322.
Yan, Bingjie & Tang, Xiangyan & Liu, Boyi & Wang, Jun & Zhou, Yize & Zheng, Guopeng & Zou, Qi & Lu, Yao & Wenxuan, Tu. (2020). An Improved Method of COVID-19 Case Fitting and Prediction Based on LSTM.
Ying, X. (2019). An overview of Overfitting and its solutions. Paper presented at the Journal of Physics: Conference Series.
Yudistira, N. (2020). COVID-19 growth prediction using multivariate long short term memory. arXiv preprint arXiv:2005.04809.
Zou, D., Wang, L., Xu, P., Chen, J., Zhang, W., & Gu, Q. (2020). Epidemic Model Guided Machine Learning for COVID-19 Forecasts in the United States. medRxiv.