A Deep Neural Networks Model for Intrusion Detection in UNSW-NB15 Dataset
Rishabh Sharma, Sakshi Sobti · 2024
The data center and cloud computing systems integration have taken over more and more central tasks in our daily activities. Although it adds more commodes through more linkages, which also implies possibilities of breaches outside human control, cybersecurity is, therefore, an essential instrument to minimize the distortion of information. In that case, the design of a scalable and elastic IDS system would be recommended. It is based on the machine learning (DL frameworks). Such systems, however, use DNNs to attempt to see and the marking, tag and sort the network attacks whether these ones are foreseen or hazy with zero-day attack patterns included. These recordings provide the basis for the automatic implementation of a targeted attack and attack prevention with close to zero chances for success. Experiments are conducted using the UNSW-NB15 dataset that demonstrates similar behaviors to those currently observed on networks with attacks artificially introduced into it.