Year: 2023
Type: article
Abstract: Network intrusion detection is crucial because of how it affects several communication and security areas, to detect network intrusion is a challenging task. Additionally, network intrusion detection is arduous work because to train modern machine learning models it takes a huge number of data to recognise an intrusion into the network. Recently, in order to identify network intrusions various methods proposed. Though they face significant difficulties because novel threats are always emerging that more older systems cannot detect. The goal of this paper is to design a network intrusion detection system (NIDS) by comparing several algorithms. Considering the relationships between the features, the optimum features are chosen from the dataset. In our research, we use K-means, Decision Tree, Random Forest, KNNs, and MLP for comparison. By monitoring the network traffic is classified into attacks and normal traffic. The experimental outcomes demonstrated that our suggested technique accurately detects various network intrusion types with on the KDD Cup99 dataset. Our proposed system also helpful for research and application areas involving network security.
Institution: Shri Jagdishprasad Jhabarmal Tibrewala University
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7216941069
