Deep Learning Approaches for Intrusion Detection System

Abstract

Nowadays, a large amount of confidential data transmitted through various devices like mobile phones and computers, and with each new innovation in computer network technology, security issues arise more frequently, and it has become difficult to ignore this issue because intruders generate new attacks every day, we must safeguard our data from them, and Intrusion Detection Systems (IDS) plays major role in that. As a result, the enhanced Intrusion Detection System was created (IDSs). Conventional intrusion detection technologies suffer from a high false-positive rates and false-negative rates, poor accuracy, and a slew of other issues. One of the issues with IDS is detecting new attacks and analyzing vast amounts of data. Deep learning is rapidly growing field that has the potential to deal with massive amounts of data. Deep learning has also been used to complete a variety of challenging projects. Deep learning is capable of governing large amounts of data and has demonstrated effective results in this area. This method is aided by its mechanism, which allows it to process enormous amounts of data quickly and accurately. Deep learning in IDS, on the other hand, has significant drawbacks. The proposed research paper will highlight how deep learning is applied in the Intrusion Detection System to improve performance, as well as a related study.