A Deep Learning Approach for Network Intrusion Detection Using a Small Feature Vector
International Journal of Advance Research and Innovation · 31.03.2026 · 14 (1) · 13-20
DOI: 10.69996/ijari.2026003 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advance Research and Innovation, online erschienen am 31.03.2026. Verfasst von Sudiksha Thatipelli.
Abstract (Originalsprache)
The increased development of computer networks and internet-related services has posed serious concerns regarding efforts to secure the systems against increasingly advanced cyberattack. The traditional network intrusion detection systems are more likely to operate on a large feature set and elaborate rule-based implementation, which imposes high computation demands and reduces efficiency in real-time implementation. As a way of overcoming these limitations, the paper proposes a deep learning-based network intrusion detection approach over a lean but effective feature vector. The given strategy will select the most informative network traffic attributes and eliminate redundant and insignificant attributes, thereby achieving dimensionality reduction. The outcome of this optimization is reduced memory and faster training and increased detection. The more complex patterns are automatically learnt using a deep learning model and the smaller set of features is used to effectively identify normal and malicious network traffic. The outcomes of the experiment have indicated that the proposed approach offers a high intrusion detection rate at significantly reduced computation time in comparison with the traditional model which has an enormous feature vector. Better still, the system is capable of picking up various types of network intrusions which include denial-of-service attacks, reconnaissance efforts, and unauthorized access. Overall, the proposed solution is a scalable, efficient, and reliable solution to the present-day systems of intrusion detection in real time over networks
Zitieren
APA
Thatipelli, S. (2026). A Deep Learning Approach for Network Intrusion Detection Using a Small Feature Vector. International Journal of Advance Research and Innovation, 14(1), 13-20. https://doi.org/10.69996/ijari.2026003
Vancouver
Thatipelli S. A Deep Learning Approach for Network Intrusion Detection Using a Small Feature Vector. International Journal of Advance Research and Innovation. 2026;14(1):13-20. doi:10.69996/ijari.2026003
BibTeX
@article{Thatipelli2026,
author = {Thatipelli, Sudiksha},
title = {A Deep Learning Approach for Network Intrusion Detection Using a Small Feature Vector},
journal = {International Journal of Advance Research and Innovation},
year = {2026},
volume = {14},
number = {1},
pages = {13--20},
doi = {10.69996/ijari.2026003}
}
EUVIMED