A REVIEW ON INCREASEING THE ACCURACY OF INTRUSION DETECTION SYSTEM (IDS)

Abstract

: In more recent times, there has been an increase in the number of people using computers, as a result of which there is a widespread use of the Internet. The use of the Internet enables hackers to access computers using new, more sophisticated, and more complex forms of attacks, to protect computers from them Intrusion Detection System (IDS) is used, which is trained with few machine learning algorithms along with datasets. The datasets used are collected over a period of time in some networks and usually contain up-to-date data. Furthermore, they are unbalanced and incapable of storing adequate data for all forms of attacks. These uneven and outdated datasets reduce the effectiveness of current IDSs, especially for attacks that are rarely encountered. In this paper, Using Random Forest, Linear Discriminant Analysis, K-Nearest Neighbor, Decision-Tree, Adaboost, and Gradient Boosting algorithms, we suggest six machine-based IDSs. To make IDS more logical, an up-to-date security database, CSE-CIC-IDS2018, is being used in place of older and more widely used datasets. The selected database is also not balanced. As a result, the rate of inequality in the dataset is reduced using a data model called the Synthetic Minority Oversampling Technique (SMOTE) to improve the reliability of the system based on the types of attacks and to eliminate unreliable access and false alarms. Data processing is done for small classes, and their numbers increase to medium data size in this way. Experimental results have shown that the proposed method significantly increases the acquisition rate of the attacks that are rarely encountered.