Deep Learning for Intrusion Detection

13A Deep Learning Approach for the Detection of Zero-day Attacks

Aamir S. Ahanger¹, Asra Khanam¹, Faheem Syeed Masoodi², and Bilal Ahmad Pandow³

¹Department of Computer Science, University of Kashmir, Srinagar, Jammu and Kashmir, India²School of ICT, Bahrain Polytechnic, Kingdom of Bahrain³Business Faculty, Bahrain Polytechnic, Kingdom of Bahrain

13.1 Introduction

Privacy and confidentiality are closely related concepts, both aiming to protect sensitive data from unauthorized access. Confidentiality measures ensure that only authorized individuals can access specific data, often utilizing encryption and access control mechanisms to safeguard information [1]. Integrity, on the other hand, focuses on maintaining the consistency, accuracy, and reliability of data throughout its lifecycle. Techniques such as hashing and cyclic redundancy checks (CRC) are commonly employed to verify that data has not been altered during storage or transmission [2]. Meanwhile, availability ensures that systems and networks remain accessible to legitimate users at all times. Organizations implement strategies like fail-safe systems and distributed architectures to guarantee continuous access to critical resources [3]. In the context of cybersecurity, integrity is compromised when malicious entities, unauthorized programs, or illegal activities infiltrate a system or network, potentially causing harm. Cybersecurity, though not universally defined, can be described as a comprehensive set of tools, techniques, ...