From Bricks to Bytes: The AI Transformation of Cybersecurity Firewalls

Due to the frequent occurrence of network threats, robust intrusion detection systems (IDS) are essential for safeguarding digital data. Traditional rule-based IDS often struggle to address emerging cyber threats, underscoring the necessity for adaptable solutions. This study presents a Next Generation Firewall (NGFW), a firewall variant that incorporates machine learning (ML) and deep learning (DL) to enhance IDS capabilities. By leveraging machine learning techniques such as Support Vector Machines (SVM), Random Forests, and Deep Neural Networks (DNN), along with deep learning models like Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN), the NGFW greatly enhances its threat detection accuracy and effectiveness. We conduct a thorough review of current literature to evaluate the efficacy of ML and DL in IDS. Our methodology involves systematically collecting and preparing data, selecting relevant features, and assessing model performance through metrics like accuracy, precision, and recall. Results indicate that CNNs are particularly effective in identifying malicious activities while maintaining a low false-positive rate. Further real-world testing of the NGFW enhances the reliability of the classification report, which demonstrates impressive accuracy. Each class achieved precision, recall, and F1-scores exceeding 97%, resulting in an overall accuracy of 97.29%. This highlights the potential for further research into intelligent cyber defense solutions, showcasing how ML and DL can contribute to the development of more resilient and adaptive security systems.