AI-Based Network Intrusion Detection System Using Machine Learning
International Scientific Journal of Engineering and Management · 13.06.2026 · 05 (06) · 1-6
DOI: 10.55041/isjem07881 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Scientific Journal of Engineering and Management, online erschienen am 13.06.2026. Verfasst von Babitha S und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Abstract—Nevertheless, the web continues to grow exponentially. Cloud tools are already in charge everywhere. but loopholes become bigger to a more cyber menacing attack. The ancient defenses are dependent on predetermined rules. They are unsuccessful in confronting new styles of attack. Such systems also fail to recognize numerous innocent behaviors. Therefore, there was a need to have a better approach. This is another method with intelligent algorithms that. The intrusion detection system tools are typically applicable to computers and are significant in ensuring that networks are not stolen by cyber thieves. These systems are machine learning and opt learning systems. It is not based on a body of rules but on patterns that are observed with regard to historical events. Decision making is done using machine learning methods. Testing is also done using real internet traffic. A data set that consists of harmful and harmless actions, referred to as CICIDS2017, was used in testing. Actual internet breakages influence the manner in which situations are tested. Moves that are not harmful and moves that are dangerous can be spotted. Close correspondence with real world events determines whether predictions can be maintained. In the starting steps such as tidying data are added to the flow noise drops out, missing spots fill introits shift form, uneven groups balance. Characteristics readjust to make learning move forward without destroying itself. After it has completed learning on the cleaned data sets, it applies a reduced interface which real time identifications of dangers normally need. Considering the figures, it is better than the existing online safety applications in terms of correctness, attention to details, and accuracy. One of the things that are known to evolve over some time is that in order to evolve, clear systems become stronger and after that surpass previous strategies of securing web connections. Pattern learning is also done through data flows to learn various patterns that are identified by the algorithm. The system becomes more apparent due to abnormal behaviour when the normal models are run. The system has a set of classifiers that define what belongs and what does not belong. [1], [6]. Keywords-Intrusion Detection System, Cybersecurity, Machine Learning, Network Traffic Analysis, Anomaly Detection, CICIDS2017 Dataset, Classification Algorithms, Model Evaluation, Security Monitoring
Zitieren
APA
S, B., M, S. N., & S, M. A. (2026). AI-Based Network Intrusion Detection System Using Machine Learning. International Scientific Journal of Engineering and Management, 05(06), 1-6. https://doi.org/10.55041/isjem07881
Vancouver
S B, M SN, S MA. AI-Based Network Intrusion Detection System Using Machine Learning. International Scientific Journal of Engineering and Management. 2026;05(06):1-6. doi:10.55041/isjem07881
BibTeX
@article{S2026,
author = {S, Babitha and M, Safrun Niyaaz and S, Mohamed Ashik},
title = {AI-Based Network Intrusion Detection System Using Machine Learning},
journal = {International Scientific Journal of Engineering and Management},
year = {2026},
volume = {05},
number = {06},
pages = {1--6},
doi = {10.55041/isjem07881}
}
EUVIMED