Network Intrusion Detection Using Supervised Machine Learning Technique with Feature Selection

In the increasingly digital world, ensuring the security of computer networks has become a crucial task. This paper introduces a machine learning-based solution to detect potential threats in network activity. By comparing the performance of two supervised learning models-Artificial Neural Networks (ANN) and Support Vector Machines (SVM)-alongside feature selection techniques, we found that ANN, when integrated with wrapper-based feature selection, yielded the highest accuracy on the NSL-KDD dataset. Our findings support the effectiveness of machine learning methods, particularly with refined feature inputs, in building robust and adaptable Network Intrusion Detection Systems (NIDS).