Methodical Applications For Cybersecurity Using Deep Learning Techniques
R Geetanjali, S. Jagannatha, Mudligiriyappa Niranjanamurthy, P. Dayananda · Solid State Technology · 2020
The research and advances in the field of Machine Learning (ML) have produalgorithms and technologies for improving security solutions that help in identifying anddecisively dealing with security threats. On the contrary, the brighter side of all theseadvancements is making it possible for cybercriminals to use the very same knowledge incrafting and launching bigger and more sophisticated cyber-attacks. This research paperfocuses mostly on the literature survey of Machine Learning (ML) and data learningtechniques for cybersecurity. Machine Learning is performing a task with the given data asinput to make its performance progressively better over time. Some of the MachineLearning and Deep Learning (DL) methods are explained and how they are relevant in thefield of Cyber Security. With increasing digitization globally, security concerns are alsogrowing at an alarming rate. The need for dynamic and progressed security technologiesand procedures to counter the complicated nature of cyber-attacks becomes imperative.This paper is intended to sensitize researchers desiring to begin their work in the field ofML or DL, and Cyber Security. Some references also have been made by citing distinctworks and some valuable examples are provided as to how cyber problems are oftentackled by ML. This paper specifically discusses both defensive and offensive behavior oncyber-attacks targeted at ML models. Finally, applications like malicious code, behavior ofnetwork traffic, malware system call sequence, and intrusion detection are also discussed.