A Survey on Cyberattack Mitigation Techniques: DDoS, Spoofing, and Ransomware Detection Using ML/DL

In the modern world, cyberattacks are emerging as a significant threat to information systems.Still, they target information networks, information devices, and information applications to steal, modify, or destroy sensitive information.These traditional methods of mitigation namely signature-based as well as rule-based detection are applicable in detecting and removing the known types of attacks, but not the zero-day attacks, polymorphic malware and wide traffic.The constantly evolving threats, including Distributed Denial of Service (DDoS), spoofing and ransomware, reiterate the need to have more adaptable defences.To overcome these impediments, cybersecurity systems are adopting machine learning (ML) and deep learning (DL).ML methods are more effective at identifying anomalies and preventing intrusion, but DL models can learn features automatically and classify complex threats effectively.Such intelligent methods are typically applied in anomaly detection and malware detection, phishing prevention, and monitoring network intrusions.The solutions provided through data-driven intelligence make use of both ML and DL to provide adaptive, scalable, and proactive mitigation against cyberattacks, which is a great improvement to the mitigation of cyber threats as compared with traditional defences because of the ever-evolving and dynamic nature of the cyber threat environment.