The current rise in hacking and computer network attacks throughout the world has heightened the demand for improved intrusion detection and prevention solutions. Cyberattacks and cybersecurity risks have skyrocketed with new technologies such as cloud computing, fog computing, edge computing, and the Internet of Things (IoT). These assaults are capable of infiltrating computer network-related environments, cloud-based services, and social networks, causing financial and reputational damage. Network intrusion detection systems play a significant role in every computer network defense system as they are used to detect and prevent malevolent activities. Netflow analysis needs efficient deep learning methods to perform real-time traffic analysis, aided by analysis of historical network traffic using traditional machine learning methods, hybrid architectures, and novel computing paradigms, such as federated learning and granular computing.

With the increasing volume of network traffic in high-speed networks, research on intelligent network traffic analysis is vital for network management and security. Deep learning specifically has been studied to solve many typical netflow analysis tasks, which have proven to be a potential general framework for analyzing network traffic. Intrusion detection systems are critical in identifying abnormalities and assaults on the network, which have grown in size, stealth, and pervasiveness. Artificial neural networks have been used in anomaly detection to determine if data behavior is normal or aberrant. With reasonable performance, this network can identify both known and unknown threats. Deep neural networks are capable of self-learning and detection of previously unknown types of network attacks, in contrast to existing systems based on signature analysis. Another advantage of neural networks is their ability to detect unknown attacks (zero-day attacks), functioning in a noisy environment, preserving operability with incomplete or distorted data, forecasting user behavior, and the emergence of new attacks.

This Special Issue aims to provide a state-of-the-art overview of novel emerging deep learning architectures and models for network intrusion and malware detection. Original research articles with a focus on both practical as well as on theoretical topics and problems, as well as review articles, are welcome.

Potential topics include but are not limited to the following:

• Artificial intelligence methods for discovering vulnerabilities and malware in wireless sensor networks and Internet of Things devices
• Artificial intelligence methods for preventing new cyberattacks
• Convolutional neural network architectures for netflow analysis and anomaly detection
• Deep learning-based solutions for low resource Internet of Things devices
• Deep learning methods to track malicious behavior and hybrid attacks in social networks
• Hybrid models, ensemble learning, and federated learning for cybersecurity
• Netflow analysis for edge, fog, and mist computing devices
• Network intrusion and malware detection using deep learning
• Novel ideas, algorithms, models, frameworks, and systems for cyberattack prediction and mitigation
• One-shot and few-shot learning for cyberattack recognition
• Real-time botnet and malware detection
• Zero-day attack recognition using deep learning methods