Next-Generation Systems and Secure Computing

15Deep Learning Algorithms for Detecting Network Attacks—An Overview

R. Mythili^* and A.S. Aneetha

Department of Computer Science, Vels Institute of Science, Technology & Advanced Studies (VISTAS), Pallavaram, Chennai, Tamil Nadu, India

Abstract

The current era of the digital and communication field is driven by Deep Learning approaches, a cutting-edge artificial intelligence technique, with a limitless array of applications that learn and evolve on their own by analyzing the environment. A classification model developed using deep learning algorithms can identify any type of data from various sources, such as images, text, and audio. The models were trained using different types of multilayer network topologies with large amounts of labeled and unlabeled data. Deep learning algorithms have evolved based on data and model outcomes used in news aggregation, fraud detection, image recognition, healthcare, and child development. Cybercrime is a rapidly changing field, and deep learning is crucial for intrusion detection systems (IDS) and malware detection systems. These systems monitor network traffic and identify suspicious activities, enabling early detection and control of cyber-attacks. Deep learning algorithms trained on large datasets improve the attack response speed, with future architectures utilizing large amounts of raw data for cyber security systems and smart automation scenarios. This chapter explores the use of deep learning algorithms in cyber security, specifically ...