A fast learning network with improved particle swarm optimization for intrusion detection system

In current days the intrusion detection systems (IDS) have several shortcomings such as high rates of false positive alerts, low detection rates of rare but dangerous attacks, and the need for a constant human intervention and tuning. Daily, there are reports of incidents such as major ex-filtration...

Full description

Saved in:

Bibliographic Details
Main Author:	Ali, Mohammed Hasan
Format:	Thesis
Language:	English
Published:	2019
Subjects:	QA76 Computer software
Online Access:	http://umpir.ump.edu.my/id/eprint/29922/1/A%20fast%20learning%20network%20with%20improved%20particle%20swarm%20optimization%20for%20intrusion%20detection%20system.wm.pdf
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-ump-ir.29922
record_format	uketd_dc
spelling	my-ump-ir.299222023-01-26T04:13:29Z A fast learning network with improved particle swarm optimization for intrusion detection system 2019-04 Ali, Mohammed Hasan QA76 Computer software In current days the intrusion detection systems (IDS) have several shortcomings such as high rates of false positive alerts, low detection rates of rare but dangerous attacks, and the need for a constant human intervention and tuning. Daily, there are reports of incidents such as major ex-filtration of data for the purposes of stealing identities, credit card numbers, and intellectual properties, as well as to take control of network resources. Machine learning approaches have been widely used to increase the effectiveness of intrusion detection platforms. While some machine learning techniques are effective at detecting certain types of attacks, there are no known methods that can be applied universally and achieve consistent results for multiple attack types. This situation makes the detection of cyber-based attacks on computer networks a relevant and challenging area of research. The Fast Learning Network (FLN) is one of the new machine learning algorithms that are easy to implement, computationally efficient, and with excellent learning performance characteristics. However, the internal power parameters (weight and basis) of FLN are initialized at random, causing the algorithm to be unstable. In this work, a new cooperative multi-swarm scheme called multi-swarmoptimization (MRPSO) which is inspired by the human social behavior was proposed for the interaction of several PSO groups while searching for the best parameters values of PSO. The focus of this research is on the development of a model that can optimize the initial parameters of FLN based on MRPSO to obtain an optimal set of initial parameters for FLN, thus, creating an optimal FLN classifier named as MRPSO-FLN which can improve the efficacy of network intrusion on data sets that contain instances of multiple classes of attacks. These methods were tested on NSL-KDD intrusiondetection datasets and the results indicate that the proposed approaches used in the system performed well in large dataset processing. In these experiments, it was demonstrated that the FLN optimization method achieved 0.9964 which is a higher accuracy than most of the existing paradigms for classifying network intrusion detection data. 2019-04 Thesis http://umpir.ump.edu.my/id/eprint/29922/ http://umpir.ump.edu.my/id/eprint/29922/1/A%20fast%20learning%20network%20with%20improved%20particle%20swarm%20optimization%20for%20intrusion%20detection%20system.wm.pdf pdf en public phd doctoral Universiti Malaysia Pahang Faculty of Computer System & Software Engineering Mohamad Fadli, Zolkipli
institution	Universiti Malaysia Pahang Al-Sultan Abdullah
collection	UMPSA Institutional Repository
language	English
advisor	Mohamad Fadli, Zolkipli
topic	QA76 Computer software
spellingShingle	QA76 Computer software Ali, Mohammed Hasan A fast learning network with improved particle swarm optimization for intrusion detection system
description	In current days the intrusion detection systems (IDS) have several shortcomings such as high rates of false positive alerts, low detection rates of rare but dangerous attacks, and the need for a constant human intervention and tuning. Daily, there are reports of incidents such as major ex-filtration of data for the purposes of stealing identities, credit card numbers, and intellectual properties, as well as to take control of network resources. Machine learning approaches have been widely used to increase the effectiveness of intrusion detection platforms. While some machine learning techniques are effective at detecting certain types of attacks, there are no known methods that can be applied universally and achieve consistent results for multiple attack types. This situation makes the detection of cyber-based attacks on computer networks a relevant and challenging area of research. The Fast Learning Network (FLN) is one of the new machine learning algorithms that are easy to implement, computationally efficient, and with excellent learning performance characteristics. However, the internal power parameters (weight and basis) of FLN are initialized at random, causing the algorithm to be unstable. In this work, a new cooperative multi-swarm scheme called multi-swarmoptimization (MRPSO) which is inspired by the human social behavior was proposed for the interaction of several PSO groups while searching for the best parameters values of PSO. The focus of this research is on the development of a model that can optimize the initial parameters of FLN based on MRPSO to obtain an optimal set of initial parameters for FLN, thus, creating an optimal FLN classifier named as MRPSO-FLN which can improve the efficacy of network intrusion on data sets that contain instances of multiple classes of attacks. These methods were tested on NSL-KDD intrusiondetection datasets and the results indicate that the proposed approaches used in the system performed well in large dataset processing. In these experiments, it was demonstrated that the FLN optimization method achieved 0.9964 which is a higher accuracy than most of the existing paradigms for classifying network intrusion detection data.
format	Thesis
qualification_name	Doctor of Philosophy (PhD.)
qualification_level	Doctorate
author	Ali, Mohammed Hasan
author_facet	Ali, Mohammed Hasan
author_sort	Ali, Mohammed Hasan
title	A fast learning network with improved particle swarm optimization for intrusion detection system
title_short	A fast learning network with improved particle swarm optimization for intrusion detection system
title_full	A fast learning network with improved particle swarm optimization for intrusion detection system
title_fullStr	A fast learning network with improved particle swarm optimization for intrusion detection system
title_full_unstemmed	A fast learning network with improved particle swarm optimization for intrusion detection system
title_sort	fast learning network with improved particle swarm optimization for intrusion detection system
granting_institution	Universiti Malaysia Pahang
granting_department	Faculty of Computer System & Software Engineering
publishDate	2019
url	http://umpir.ump.edu.my/id/eprint/29922/1/A%20fast%20learning%20network%20with%20improved%20particle%20swarm%20optimization%20for%20intrusion%20detection%20system.wm.pdf
_version_	1783732125050077184

A fast learning network with improved particle swarm optimization for intrusion detection system

Similar Items