Hybrid Feature Selection (RFEMI) Techniques and Intrusion Detection Systems for Web Attacks Detection Using Supervised Machine Learning Algorithms.
Ibrahim Abobaker · 2023
In the realm of business, it is crucial to establish robust security mechanisms for identifying web attacks. Advanced Intrusion Detection Systems (IDSs) can effectively fortify network security by operating at the network’s perimeter. However, the presence of a vast number of features within the system presents a significant challenge in accurately detecting web attackers. To address this issue, a technique known as Recursive Feature Elimination and Mutual Information (RFEMI) is proposed. This technique aims to select the most essential features while eliminating duplicates, thereby simplifying attack detection and reducing computational time. The study conducted experiments to assess the effectiveness of the proposed feature selection technique in detecting web attacks using various Machine Learning Algorithms (MLAs) such as Decision Tree Classifier (DTC), XGB Classifier (XGB), Gradient Boosting Classifier (GBC) and K-Nearest Neighbor (KNN) for Network-based Intrusion Detection Systems (NIDS). The results demonstrate that the proposed system can accurately identify web attacks using the CICIDS-2017 dataset. Among the classifiers, the DTC classifier exhibited the highest accuracy at 0.9961, with a False Positive Rate (FPR) of 0.054.