Published Paper Details:

Cyber analysts find it more and more challenging to efficiently monitor the volume, velocity, and diversity of data produced as the number of Internet-connected equipment grows. Cyber security tactics that rely on signatures are unlikely to deliver the performance needed to find new attack vectors. The development of advanced attack techniques that can evade detection by present security systems is also made possible by technical advancements. We require cutting-edge tools and technologies to identify, investigate, and take prompt action on new assaults and threats as the cyber threat scenario gets worse. To identify various sorts of attacks, a cyber-security system must be constructed. The use of various intelligence algorithms in cyber security enabled the detection and analysis of attacks on computer networks. Artificial intelligence, machine learning, and deep learning algorithms are used in cyber security to take the best feature representation possible out of a large data set. This has been used in a number of cyber security scenarios, including the analysis, prediction, and detection of attacks. The purpose of this work is to analyze cyber security attack datasets using clever methods. Additionally, it offers a thorough comparing of algorithm effectiveness and field application to explain the advantages of network protection optimization methods. We analyze the key traits of reflective deep learning approaches used in cyber security application domains, we bring the recent developments in deep learning, and we offer an insight of resources required like a systematic framework and appropriate datasets before analytically and comparably assessing state-of-the-art services from the research. We point out the shortcomings of the examined works and present a picture of the current problems in the field, offering helpful advice and best practices for academics and developers tackling relevant issues. Finally, we identify current research directions and pain issues that need to be addressed.