MBC-UBEM-IDS: A Two-Step Intrusion Detection System for Imbalanced Network Traffic

Abstract

Intrusion Detection System is a type of tool capable of effectively identifying malicious network traffic. However, existing intrusion detection methods face the challenge of class imbalance, making it difficult to correctly classify the minority class. To address this issue, a two-step intrusion detection system is proposed, employing a multi-branch convolutional neural network in the first step and combining undersampling algorithm with One-Versus-Others strategy in the second step. The performance of the proposed method has been evaluated on the NSL-KDD dataset, demonstrating superior performance compared to other methods.