Intrusion Detection and Prevention System for DDoS Attacks using Hybrid CNN Architecture

Machine-based intrusion detection systems are crucial for network protection and data security in this age of widespread Internet connectivity and consistent growth in data quantities. However, current methods of both shallow and deep learning fall short when trying to capture the intricate incursion patterns associated with data transmission. To combat this, we offer a large data-based hierarchy deep learning system to further improve the effectiveness of IDS-based artificial intelligence. Both network traffic as well as content information may be captured by the proposed system thanks to its behavioral and content-functional features. Each framework deep learning model is dedicated to understanding the data structure of a specific cluster. For this reason, we advocated the use of a convolutional Neural Network (CNN) using a Min-Max optimization approach. When compared to earlier single-learning models, the suggested method increases the categorization rate of disruptive attacks. Incorporating a larger number of devices also speeds up the system's model-building process.