An empirical study of Machine Learning Approaches for Intrusion Detection System

Authors

  • Paryana Tahiri , Dr. Sonia , Dr. Pankaj Jain

Abstract

IDS (Intrusion Detection System) is a software application that keeps track of malicious activities
in a network system. An attacker may concession the CIA Triad (Confidentiality, Integrity and Availability)
of network resources. While there exist some sorts of the violation. Thus, there arises a dire need to study
the current scenario of harmful network attacks, one possible solution for protection over vulnerability is
utilizing the IDS system with machine learning techniques. Prominent researchers have worked and
implemented various machine learning techniques in order to build a smart IDS. In this paper, we will do a
thorough review of several previous research studies to find out the rapid challenges existed in the field of
IDS. While, how the machine learning techniques has been used for different scenario of data mining,
feature selection and classification approaches used for IDS. As well as, the ongoing research gap,
challenges and at the same time as a result of the review study the effective outcomes from the previous
research experiments has been pointed. In order to offer guideline for the IDS improvement in the practical
world

Published

2020-01-31

Issue

Section

Articles