International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
|
Volume 186 - Issue 50 |
Published: November 2024 |
Authors: Khaled Bin Showkot Tanim, Mahadi Hasam Parash, Md Shadman Soumik, Mohammed Shakib |
![]() |
Khaled Bin Showkot Tanim, Mahadi Hasam Parash, Md Shadman Soumik, Mohammed Shakib . Enhanced Network Anomaly Detection Using Convolutional Neural Networks in Cybersecurity Operations. International Journal of Computer Applications. 186, 50 (November 2024), 13-25. DOI=10.5120/ijca2024924224
@article{ 10.5120/ijca2024924224, author = { Khaled Bin Showkot Tanim,Mahadi Hasam Parash,Md Shadman Soumik,Mohammed Shakib }, title = { Enhanced Network Anomaly Detection Using Convolutional Neural Networks in Cybersecurity Operations }, journal = { International Journal of Computer Applications }, year = { 2024 }, volume = { 186 }, number = { 50 }, pages = { 13-25 }, doi = { 10.5120/ijca2024924224 }, publisher = { Foundation of Computer Science (FCS), NY, USA } }
%0 Journal Article %D 2024 %A Khaled Bin Showkot Tanim %A Mahadi Hasam Parash %A Md Shadman Soumik %A Mohammed Shakib %T Enhanced Network Anomaly Detection Using Convolutional Neural Networks in Cybersecurity Operations%T %J International Journal of Computer Applications %V 186 %N 50 %P 13-25 %R 10.5120/ijca2024924224 %I Foundation of Computer Science (FCS), NY, USA
Network anomaly detection is critical for preserving cybersecurity and safeguarding sensitive data. Traditional approaches sometimes struggle with the complexity and amount of current network traffic. This research provides an upgraded network anomaly detection method utilizing convolutional neural networks (CNNs). Leveraging the BoT-IoT dataset, this paper utilize feature selection strategies based on entropy and correlation to develop a robust CNN feature matrix. The model showed considerable gains in identifying abnormalities, with a high accuracy rate of 96%. The application of the system in both offline and online modes illustrates its relevance in real-world cybersecurity operations. Detailed assessments, including training and testing timeframes, indicate the system's efficiency and efficacy. Future work will concentrate on increasing the dataset, incorporating additional deep learning models, and boosting real-time detection capabilities.