隨著時代改變資訊安全攻擊手法逐漸多樣化，透過高速網路及高頻寬可達成多種攻擊含殭屍電腦、網路挖礦、APT進階持續攻擊、DDoS分散式阻斷服務攻擊等。由此可知攻擊與網路為相對關係，以目前攻擊手法最普遍也最難防禦屬分散式阻斷服務攻擊(Distributed Denial of Service, DDoS)，要有效抑制及阻擋困難度很高，利用已被入侵的電腦進行遠端控制(殭屍電腦)向攻擊目標發出大量封包導致系統服務癱瘓，現階段無資訊安全設備可全面防禦，只可透過流量清洗設備進行封包清洗以恢復系統服務。
承上，若分散式阻斷服務攻擊(Distributed Denial of Service, DDoS)為目前熱門的資訊安全攻擊手法之一，為何不探討該攻擊手法的過去及現在有無改變，五年前與五年後人們所探討的分散式阻斷服務攻擊(Distributed Denial of Service, DDoS)代表意義有什麼差異，網路攻擊手法包含非常多因素，如系統資源、硬體設備、網路封包、資料加密、安全傳輸等，因此亦希望透過某種方式分析分散式阻斷服務攻擊之主要議題為何 ?
本次希望透過機器學習的方式，將資訊安全之阻斷式服務攻擊手法議題進行分析，透過在線學習(Online Learning)、非監督式學習(AutoEncoder)、主題模型(Topic Model)進行字詞與主題的研究，同時制作研究領域字典含文章抓取、文字預處理、轉檔及過濾、正規化等一併探討，以達本研究目標。

The method of information security attack has become increasingly complex today.
In the high-speed and high-bandwidth network environment, among cyber security attack methods such as zombie computers, malicious mining, APT (advanced continuous attacks), and DDoS (distributed denial of service), the most difficult to defend is the DDoS. It is extremely difficult to effectively restrain or block. The attacker leverages malicious software to remotely control the invaded computers, and creates a botnet to send a large number of request packets, causing the target system services to be out of services. No information security equipment can effectively defend DDoS attack, and only traffic scrubbing can be performed through the traffic scrubbing device to mitigate malicious attack traffic, and achieve the purpose of restoring system services.
The DDoS is one of the current popular information security attack threats, and related issues have evolved. what is the difference in terms of Distributed Denial of Service (DDoS) attacks five years ago and now? Network attack methods include many factors, such as system resources, hardware equipment, network packets, data encryption, secure transmission, etc. We hope to analyze what are the topics discussed in recent years about blocked service attacks on information security.
This paper will analyze the topic of distributed denial of service attacks on information security through machine learning. We use Online Learning, Auto Encoder and Topic Model methods to conduct research between words and topics, at the same time, we discuss the dictionary generation including article capture, text pre-processing, file conversion and filtering, distance, etc. in order to achieve the goal of this research.

論文審定書 i
誌謝 ii
摘要 iii
Abstract iv
目錄 v
圖次 vi
表次 viii
第一章、 研究背景、動機及目的 1
第一節、 研究背景 1
第二節、 研究動機 3
第三節、 研究目的 4
第二章、 文獻探討 5
第一節、 教育系資安通報平台 5
第二節、 arXiv論文平台 7
第三節、 文字分析技術探討分散式攻擊(DDoS)相關研究 8
第四節、 以自然語言處理庫(NLP)技術建置文字檔相關研究 9
第三章、 研究設計及方法 12
第一節、 研究方法 12
第二節、 研究資料 12
第三節、 資料預處理過程 14
第四節、 產生資訊安全字典集 18
第四章、 研究成果 22
第一節、 資料集說明 22
第二節、 實驗說明 24
第五章、 研究結論 33
第六章、 參考文獻 34

Bird, S. (n.d.). NLTK-Lite: Efficient Scripting for Natural Language Processing. 9.
Chen, X., Kingma, D. P., Salimans, T., Duan, Y., Dhariwal, P., Schulman, J., Sutskever, I., & Abbeel, P. (2017). Variational Lossy Autoencoder. ArXiv:1611.02731 [Cs, Stat]. http://arxiv.org/abs/1611.02731
Curtis, D. D., & Lawson, M. J. (2019). EXPLORING COLLABORATIVE ONLINE LEARNING. Online Learning, 5(1). https://doi.org/10.24059/olj.v5i1.1885
Douligeris, C., & Mitrokotsa, A. (2004). DDoS attacks and defense mechanisms: Classification and state-of-the-art. Computer Networks, 44(5), 643–666. https://doi.org/10.1016/j.comnet.2003.10.003
Feinerer, I., Hornik, K., & Meyer, D. (2008). Text Mining Infrastructure in R. Journal of Statistical Software, 25(5). https://doi.org/10.18637/jss.v025.i05
Feinstein, L., Schnackenberg, D., Balupari, R., & Kindred, D. (2003). Statistical approaches to DDoS attack detection and response. Proceedings DARPA Information Survivability Conference and Exposition, 303–314. https://doi.org/10.1109/DISCEX.2003.1194894
Haider, S., Akhunzada, A., Mustafa, I., Patel, T. B., Fernandez, A., Choo, K.-K. R., & Iqbal, J. (2020). A Deep CNN Ensemble Framework for Efficient DDoS Attack Detection in Software Defined Networks. IEEE Access, 8, 53972–53983. https://doi.org/10.1109/ACCESS.2020.2976908
Keromytis, A. D., Misra, V., & Rubenstein, D. (2004). SOS: An Architecture for Mitigating DDoS Attacks. IEEE Journal on Selected Areas in Communications, 22(1), 176–188. https://doi.org/10.1109/JSAC.2003.818807
Kivinen, J., Smola, A. J., & Williamson, R. C. (2004). Online Learning with Kernels. IEEE Transactions on Signal Processing, 52(8), 2165–2176. https://doi.org/10.1109/TSP.2004.830991
Lemme, A., Reinhart, R. F., & Steil, J. J. (2010a). Efficient online learning of a non-negative sparse autoencoder. Computational Intelligence, 6.
Lemme, A., Reinhart, R. F., & Steil, J. J. (2010b). Efficient online learning of a non-negative sparse autoencoder. Computational Intelligence, 6.
Lemme, A., Reinhart, R. F., & Steil, J. J. (2012). Online learning and generalization of parts-based image representations by non-negative sparse autoencoders. Neural Networks, 33, 194–203. https://doi.org/10.1016/j.neunet.2012.05.003
Li, J., Luong, M.-T., & Jurafsky, D. (2015). A Hierarchical Neural Autoencoder for Paragraphs and Documents. ArXiv:1506.01057 [Cs]. http://arxiv.org/abs/1506.01057
Li, Y., & Lu, Y. (2019). LSTM-BA: DDoS Detection Approach Combining LSTM and Bayes. 2019 Seventh International Conference on Advanced Cloud and Big Data (CBD), 180–185. https://doi.org/10.1109/CBD.2019.00041
Liou, C.-Y., Cheng, W.-C., Liou, J.-W., & Liou, D.-R. (2014). Autoencoder for words. Neurocomputing, 139, 84–96. https://doi.org/10.1016/j.neucom.2013.09.055
Ong, C. S. (n.d.). TOM.DESMEDT@UA.AC.BE WALTER.DAELEMANS@UA.AC.BE. 5.
Proceedings of the COLING/ACL 2006 Interactive Presentation Sessions. (n.d.). 4.
Ramos, J. (n.d.). Using TF-IDF to Determine Word Relevance in Document Queries. 4.
Rana, A., & Kamboj, A. (n.d.-a). Computer Science and Engineering/Information Technology. 54.
Rana, A., & Kamboj, A. (n.d.-b). Computer Science and Engineering/Information Technology. 54.
Tama and Rhee—Data Mining Techniques in DoSDDoS Attack Detectio.pdf. (n.d.).
Tama, B. A., & Rhee, K.-H. (n.d.). Data Mining Techniques in DoS/DDoS Attack Detection: A Literature Review. 10.
Thompson, K. (1968). Programming Techniques: Regular expression search algorithm. Communications of the ACM, 11(6), 419–422. https://doi.org/10.1145/363347.363387
Wang and Zhang—2017—DDoS Event Forecasting using Twitter Data.pdf. (n.d.).
Wang, Z., & Zhang, Y. (2017). DDoS Event Forecasting using Twitter Data. Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, 4151–4157. https://doi.org/10.24963/ijcai.2017/580
Yaar, A., Perrig, A., & Song, D. (2003). Pi: A path identification mechanism to defend against DDoS attacks. Proceedings 19th International Conference on Data Engineering (Cat. No.03CH37405), 93–107. https://doi.org/10.1109/SECPRI.2003.1199330
Yu, W., Zheng, C., Cheng, W., Aggarwal, C. C., Song, D., Zong, B., Chen, H., & Wang, W. (2018). Learning Deep Network Representations with Adversarially Regularized Autoencoders. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2663–2671. https://doi.org/10.1145/3219819.3220000