Performance evaluation for different intrusion detection system algorithms using machine learning

Intrusion is a set of operations that decide to compromise the integrity, confidentiality, and convenience of pc resources. This definition ignores the success or failure of those operations, therefore it additionally corresponds to attacks on pc systems. The distinguishing actions that mitigate the operations to compromise the integrity, confidentiality, or convenience of pc resources is termed as intrusion detection. The aim of the Intrusion Detection System (IDS) is to monitor the network system for any sort of attacks. The objectives of this project is to evaluate the performance of various intrusion detection algorithms based on machine learning. The algorithms considered are the Naive Bays Algorithm, Decision Tree Algorithm and Hybrid Algorithm for different datasets. The evaluation of of various intrusion detection algorithms for different datasets is done utilizing a set of performance metrics, including accuracy, precision, central processing unit (CPU) efficiency and execution time. The simulation results showed that the Decision Tree Algorithm achieved the highest precision rate by 95.5% within 42 seconds. For accuracy, it achieved 69.5% and can be considered good, as compared to all other algorithms. Additionally, the Decision Tree Algorithm records 22.5% in CPU utilization. For Naive Byes Algorithm it scored 90% for Precision, 65% for Accuracy and 14% for CPU Utilization. Lastly, for Algorithm, it obtained 91% for Precision76% for Accuracy and 23% for CPU.

Name dc:type.qualificationname

mphil

Level dc:type.qualificationlevel

masters

Grantor dc:publisher.institution

Universiti Tun Hussein Onn Malaysia

Year dc:date.issued

2018

Author dc:creator

• Zarir, Mustafa Nadhim

• TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television

Language dc:language

en