Web Attack Intrusion Detection System Using Machine Learning Techniques - Research paper

Aug 2, 2026

Uploaded by Oliver

Web attacks commonly target network-accessible web applications and exploit vulnerabilities such as SQL injection, XSS, and brute-force attempts. Detecting these threats effectively depends on traffic classification in intrusion detection systems (IDS). This study evaluates machine-learning approaches using the CIC-IDS2017 dataset, which includes 80 attributes of recent assaults. Three algorithms—random forests, k-nearest neighbors, and naive Bayes—are compared by accuracy, precision, and recall.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages15

Reading Time38 min

AccessFree to download

Introduction

Related work and datasets

Methodology

Dataset selection (CIC-IDS2017)

Algorithms evaluated (RF, KNN, NB)

Experimental results

Accuracy comparison

Precision and recall analysis

Conclusion

JOE International Journal of Online and Biomedical Engineering [Onli](Online-Journals.org)[ne-Jo](Online-Journals.org)[urnals](Online-Journals.org)[.org](Online-Journals.org) iJOE | eISSN: 2626-8493 | Vol. 20 No. 3 (2024) | [https://doi.org/10.3991/ijoe.v20i03.45249](https://doi.org/10.3991/ijoe.v20i03.45249) PAPER Web Attack Intrusion Detection System Using Machine Learning Techniques Mahmoud Khalid Baklizi1, Issa Atoum2, Mohammad Alkhazaleh1, Hasan Kanaker3, Nibras Abdullah4,5(􀀍), OlaA. Al-Wesabi5, Ahmed Ali Otoom6 1Department of Computer Sciences, Faculty of Information Technology, Isra University, Amman, Jordan 2Software Engineering Department, Faculty of Information Technology, The World Islamic Sciences and Education, Amman, Jordan 3Department of Cybersecurity, Faculty of Information Technology, Isra University, Amman, Jordan 4School of Computer Sciences, Universiti Sains Malaysia (USM), Gelugor, Malaysia 5Faculty of Computer Science and Engineering, Hodeidah University, Hodeidah, Yemen 6Cybersecurity and Cloud Computing Department, Faculty of Information Technology, Applied Science Private University, Amman, Jordan [nibras@usm.my](nibras@usm.my) ABSTRACT Web attacks often target web applications because they can be accessed over a network and often have vulnerabilities. The success of an intrusion detection system (IDS) in detecting web attacks depends on an effective traffic classification system. Several previous studies have utilized machine learning classification methods to create an efficient IDS with various datasets for different types of attacks. This paper utilizes the Canadian Institute for Cyber Security’s (CIC-IDS2017) IDS dataset to assess web attacks. Importantly, the dataset contains 80 attributes of recent assaults, as reported in the 2016 McAfee report. Three machine learning algorithms have been evaluated in this research, namely random forests (RF), k-nearest neighbor (KNN), and naive bayes (NB) . The primary goal of this research is to propose an effective machine learning algorithm for the IDS web attacks model. The evaluation compares the performance of three algorithms (RF, KNN, and NB) based on their accuracy and precision in detecting anomalous traffic. The results indicate that the RF outperformed the NB and KNNin terms of average accuracy achieved during the training phase. During the testing phase, the KNN algorithm outperformed others, achieving an average accuracy of 99.4916% . However, RF and KNN achieved 100% average precision and recall rates compared to other algorithms. Finally, the RF and KNN algorithms have been identified as the most effective for detecting IDS web attacks. KEYWORDS intrusion detection systems, CIC-IDS2017, machine learning, false alarms, naive bayes (NB), k-nearest neighbors (KNN), random forest (RF) 1 INTRODUCTION Despite the tremendous development of the Internet, it is still vulnerable to security issues in web-based applications. These vulnerabilities give hackers the ability to carry out various web attacks, such as SQL injection, XSS, and brute force [1] . In web attacks, hackers exploit vulnerabilities to gain unauthorized access to web Baklizi, M. K., Atoum, I., Alkhazaleh, M., Kanaker, H., Abdullah, N., Al-Wesabi, O.A., Otoom, A.A. (2024) . Web Attack Intrusion Detection System Using Machine Learning Techniques. International Journal of Online and Biomedical Engineering (iJOE), 20(3), pp. 24–38. [https://doi.org/10.3991/ijoe](https://doi.org/10.3991/ijoe). v20i03.45249 Article submitted 2023-09-24. Revision uploaded 2023-11-15. Final acceptance 2023-11-23. © 2024 by the authors of this article. Published under CC-BY. 24 International Journal of Online and Biomedical Engineering (iJOE) iJOE | Vol. 20 No. 3 (2024) Web Attack Intrusion Detection System Using Machine Learning Techniques servers, network devices, or the network itself [2] . Hence the importance of intrusion detection systems, which play a cr

FAQs

What dataset is used to evaluate web attack detection in this study?

The paper uses the Canadian Institute for Cyber Security’s CIC-IDS2017 IDS dataset, which contains 80 attributes of recent assaults.

Which machine learning algorithms are compared for intrusion detection?

Three algorithms are evaluated: random forests (RF), k-nearest neighbors (KNN), and naive Bayes (NB).

What performance results are reported for the three algorithms?

RF achieves the best average accuracy during training, while KNN reaches the highest average accuracy during testing (99.4916%). RF and KNN also show 100% average precision and recall rates.

Aug 2, 2026

Uploaded by Oliver