A Comprehensive Study ofIntelligent Techniques for Detecting Network Attacks

Abstract

Because of the growing increase and adoption of network-based communication technologies, cybersecurity has become a serious concern, especially as the number of cyber-attacks rises. In order to detect known attacks utilising signatures in network traffic, a variety of detection algorithms are deployed. Researchers have utilised several machine learning algorithms to detect network assaults without depending on signatures in recent years. The approaches have a significant false-positive rate, which is insufficient for an intrusion detection system that is ready for the market.

In this research, the author has emphasised the critical measurements and parameters in relation to massive organisational situations for protecting a large amount of data. Developers and organisations use security measures to prevent them from attaining their goals. The purpose of this research is to discover and prioritise security ways for locating and solving problems using different approaches of machine learning that have previously been used to analyse big data security. Authors are examining the priorities and overall data security using the Machine learning approach. In addition, the most relevant weight-related characteristics have been quantified. Experts will learn about the findings and conclusions that will help them improve big data security.