AI Based Intrusion Detection System for IoT Enabled Smart Industries

In the modern era of extensive data utilization, network security encounters significant threats necessitating innovative countermeasures. This research introduces a Hybrid Network Intrusion Detection System (HNIDS) that integrates Convolutional Neural Networks (CNNs) with k-Nearest Neighbors (KNN) to address the challenges of detecting intrusions in data-intensive, imbalanced environments. By leveraging CNNs for feature extraction and employing SMOTE oversampling combined with Tomek-links under sampling to balance datasets, the system achieves accurate detection of intrusion patterns. Evaluation results highlight enhanced detection accuracy compared to standalone models, showcasing the system's capability to manage complex network security scenarios effectively.