A Comprehensive Tutorial and Survey of Applications of Deep Learning for Cyber Security
Deep learning (DL), a novel research direction in machine learning (ML) field, has achieved great success in many classical artificial intelligence (AI) tasks in comparison to classical ML algorithms (CMLAs). DL architectures are relatively recent and currently wisely used for diverse Cyber Security applications. This work aims to review the state-of-the-art DL architectures in Cyber Security applications by highlighting the contributions and challenges from various recent research papers. Initially, the concepts of most popular ML algorithms and DL architectures are discussed along with the mathematical representations. Following, we review the emerging researches of DL architectures for diverse anticipated applications of Cyber Security. This include Intrusion detection, Malware and Botnet detection, Spam and Phishing detection, Network traffic analysis, Binary analysis, Insider threat detection, CAPTCHA analysis, steganography. Additionally, the importance of DL architectures are discussed for cryptography, cloud security, biometric security, smart cities specific to Internet of things (IoT) and fog computing. We discuss the importance of big data, natural language processing, signal and image processing, blockchain technology, casual theory key concepts towards Cyber Security. Finally the paper concludes with a summary of the current state-of-the-art, a critical discussion of open challenges and directions for future research additionally we propose and discuss general DL based Cyber Security system.
Cyber Security, Machine Learning, Neural Networks, Deep Learning