Paper Contents
Abstract
The growing dependence on the internet has increased the vulnerability of businesses to cyber threats. Network Intrusion Detection Systems (NIDS) are essential tools for safeguarding sensitive data and network infrastructure by identifying and flagging suspicious activities that may indicate malicious intent. This paper introduces a novel NIDS that leverages a combination of machine learning algorithms to enhance anomaly detection and classification. The system employs Support Vector Machines, K-Nearest Neighbors, Random Forest, and Logistic Regression to effectively differentiate between legitimate network traffic and a diverse range of attack types. This multifaceted approach capitalizes on the strengths of each algorithm: SVMs for robustly classifying high-dimensional data, KNN for efficient multi-class classification, Random Forests for handling complex datasets and mitigating overfitting, and Logistic Regression for interpretability and scalability. The NIDS utilizes the NSL-KDD dataset, a widely recognized benchmark for NIDS evaluation, to train and assess model performance.
Copyright
Copyright © 2024 Shrinivas Jadhav. This is an open access article distributed under the Creative Commons Attribution License.
