Analysis of Network Intrusion Detection Technique based on Machine Learning

In today's world internet usage is on the rise. Every human task is accomplished with the help of the internet. Network security is an important factor in such conditions. Maintaining network availability, integrity and confidentiality is important. This requires a network administrator to accept various types of intrusion detection systems (IDS) that help monitor network origins of unauthorized and dangerous activities. An intrusion Detection System (IDS) is a system that monitors traffic on the network for suspicious activity and it alerts the issues when that activity is available. It is a software application that scans the network or the program for malicious activity or policy violations. Any malicious or illegal engagement is usually reported to the regulator or collected through an event security information and event management system (SIEM). The SIEM system integrates output from multiple sources and uses alarm filtering methods to distinguish hazardous activity from false alarms. The presented paper gives a clear idea about what is the need for IDS and also what are the network intrusion detection techniques using machine learning. The techniques are mainly divided into two parts. first is classical artificial intelligence (AI) and the second is computational intelligence (CI). The proposed paper will explore the brief scenario about these two methods for developing effective IDS.