Deep Neural Network Based Cyber Intrusion Detection System

An Intrusion Detection System (IDS) is a critical security device used to prevent intrusions, protect networks and digital assets from malicious traffic. Considering the rapid advancement of internet technology, security of networks has become a critical concern as there is high rise in data volume due to exponential growth of internet users and thus posing the serious security challenges. Soft computing techniques have proven effective in handling such large datasets, outperforming traditional approaches and demonstrating superior performance. In this study, we propose a soft computing-based IDS, training the model using complete set of features. The system was simulated with the help of a hybrid real and modern threat activities synthesized in UNSW-NB15 dataset, and the results validate its effectiveness.