Journals / Tasarım Mimarlık ve Mühendislik Dergisi / 2021 / Cilt: 1 - Sayı: 3

DETECTION OF DDOS ATTACKS WITH MACHINE LEARNING ALGORITHMS

DDOS SALDIRILARININ MAKİNE ÖĞRENİMİ ALGORİTMALARIYLA TESPİTİ

Pages
221–232
Publication Date
2021-10-30
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The highest accuracy rate of 99% was reached in the Multi Layer Perceptron model.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:DDoS attacks are the most common type of attack among network attacks.
📊
Key Findings & Data:ROC curves and Precision, Recall, F1-Score and Accuracy metrics were used in the evaluation.
💡
Conclusion & Impact:The highest accuracy rate of 99% was reached in the Multi Layer Perceptron model.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:DDoS · Network Saldırıları · Makine Öğrenimi

Abstract

DDoS attacks are the most common type of attack among network attacks. As a result of these attacks, individuals and companies have to deal with problems such as loss of money, reputation and time. Although different methods and systems have been tried to prevent attacks, it cannot be said that the problems have been completely eliminated. One of the solutions to problems is to detect attacks early. This study aims to classify attacks with machine learning methods to detect DDoS attacks. In the study, the data in the selected data set was optimized and K-Nearest Neighbors, Multi Layer Perceptron, Support Vector Machine and Random Forest classifier models were developed. ROC curves and Precision, Recall, F1-Score and Accuracy metrics were used in the evaluation. The highest accuracy rate of 99% was reached in the Multi Layer Perceptron model.

Özet

DDoS saldırıları, network saldırıları içerisinde en sık rastlanan saldırı türüdür. Bu saldırılar sonucunda bireyler ve şirketler para, itibar ve zaman kaybı gibi sorunlarla uğraşmak zorunda kalmaktadırlar. Saldırıları önleme amaçlı farklı yöntem ve sistemler denenmekle birlikte sorunların tamamen ortadan kalktığı söylenemez. Sorunlara çözüm önerilerinden birisi saldırıların erken tespit edilmesidir. Bu çalışma; DDoS saldırılarının tespiti için saldırıların makine öğrenmesi yöntemleriyle sınıflandırılmasını amaçlamaktadır. Çalışmada, seçilen veri setindeki veriler optimize edilmiş ve K-Nearest Neighbours, Multi Layer Perceptron, Support Vector Machine ve Random Forest sınıflayıcı modelleri geliştirilmiştir. Değerlendirmede ROC eğrileri ile Precision, Recall, F1-Score ve Accuracy metriklerinden yararlanılmıştır. En yüksek doğruluk oranı olan %99’a Multi Layer Perceptron modelinde ulaşılmıştır.

🔖

Cite This Article

ASARKAYA, S., KAYNAR, O., YELMEN, İ., YILDIRIM, F., ZONTUL, M. (2021). DETECTION OF DDOS ATTACKS WITH MACHINE LEARNING ALGORITHMS. Tasarım Mimarlık ve Mühendislik Dergisi, Cilt: 1 - Sayı: 3, 221–232. https://ajindex.com/articles/855025

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (6)

  1. Aamir, M., & Zaidi, S. A., “Clustering based semi-supervised machine learning for DDoS attack classification”, Journal of King Saud University, Feb. 2019. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S131915781831067X?via%3Dihub
  2. Arivudainambi, D., Varun, K. K., Sibi, C. S., & Visu, P., “Malware traffic classification using principal component analysis and artificial neural network for extreme surveillance”, Computer Communications, vol. 147, pp. 50-57, Nov. 2019, doi: 10.1016/j.comcom.2019.08.003 Branitskiy, A., & Kotenko, I., “Hybridization of computational intelligence methods for attackdetection in computer networks”, Journal of Computational Science, vol. 23, pp. 145-156, Nov. 2017, doi: 10.1016/j.jocs.2016.07.010 Cao, Y., Wang, Z., Qian, Z., Song, C., Krisnamurhy, S., & Yu, P., “Principled Unearthing of TCP Side Channel Vulnerabilities”, presented at 2019 ACM SIGSAG Conference on Computer and Communication Security, London, UK, Nov. 11-15, 2019 DDoS attack network logs dataset, www.kaggle.com, [Online]. Available: https://www.kaggle.com/jacobvs/ddos-attack-network-logs (Erişim 1 Ocak 2021)
  3. Deka, R. K., Bhattacharyya, D. K., & Kalita, J. K., “Active learning to detect DDoS attack using ranked features”, Computer Communications, vol. 145, pp. 203-222, Sep. 2019, doi: 10.1016/j.comcom.2019.06.010 Deore, S., & Patil, A., “Survey Denial of Service classification and attack with Protect Mechanism for TCP SYN Flooding Attacks.”, International Research Journal of Engineering and Technology (IRJET), vol. 3, no. 5, pp. 1736-1739, May, 2016.
  4. Dimitrios, K., Apostolos, P., Maglaras, L., Moschoyiannis, S., Aparicio-Navarro, F., Argyriou, A., & Janicke, H., “A novel Intrusion Detection System against spoofing attacks in connected Electric Vehicles”, Array, vol. 5, March 2020, Art. no. 100013. Gavaskar, S., Surendiran, R., & Ramaraj, D., “Three Counter Defense Mechanism for TCP SYN Flooding Attacks”, International Journal of Computer Applications, vol. 6, no.6 Sep, 2010
  5. Hoo, Z. H., Candlish, J., & Teare, D., “What is an ROC curve?”, Emergency Medicine J, vol. 34, Mar. 2017, Art. no. 206735. doi: 10.1136/emermed-2017-206735 Kariyer Uygulama ve Araştırma Merkezi. “FileHandler2.ashx”. http://kariyer.istanbul.edu.tr. http://cdn.istanbul.edu.tr/FileHandler2.ashx?f=bilisimdekariyer2.pdf (Erişim 3 Ocak 2021)
  6. Kothari, N., Mahajan, R., Millstein, T., Govindan, R., & Musuvathi, M., “Finding Protocol Manipulation Attacks”, Association for Computing