Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/475793| Title: | Linear genetic programming and bees algorithm (LGP_BA) feature selection algorithm for network Intrusion Detection System |
| Authors: | Seyed Reza Hasani (P54325 ) |
| Keywords: | Intrusion Detection System Algorithm Programming Universiti Kebangsaan Malaysia -- Dissertations. |
| Issue Date: | 9-Feb-2012 |
| Description: | Network security is a serious global concern. Intrusion Detection Systems (IDS) are one of the emerging areas of Information Security research to be implemented using Soft computing techniques. Generally, intrusion detection methods are dealing with huge amount of data. Most of the existing IDSs use all the features in network packet to evaluate and look for known intrusive patterns. Some of these features are irrelevant and redundant. In addition, the selection of a subset will reduce the dimensionality of the data samples and eliminate the redundancy and ambiguity introduced by some attributes. Thus, feature selection is required to address this issue. Several feature selection Algorithms has been applied in IDS such as Rough set, Rough DPSO, LGP, MARS, SVDF, and BA. Past research show Linear Genetic Programming (LGP) algorithm has obtained high Detection Rate (DR), whilst Bees Algorithm (BA) has low False Alarm Rate (FAR) for IDS. Therefore, this research propose LGP_BA feature selection algorithm for IDSs to remain high DR and reduce FAR. Support Vector Machine (SVM) is the most popular classifier to provide potential solutions for the IDSs problems and to determine whether the derived feature set is well selected or not. The performance of the proposed method is achieved using the KDD cup‟99 benchmark dataset. In this work four sample datasets containing 4000 random records are used. Experimental results show that the LGP_BA method improves 1% the DR and reduces 6% FAR in comparison with LGP. The research also compared with the other past algorithms based on ROC.,Certification of Master's/Doctoral Thesis" is not available |
| Pages: | 112 |
| Call Number: | TK5105.59.H374 2013 3 tesis |
| Publisher: | UKM, Bangi |
| URI: | https://ptsldigital.ukm.my/jspui/handle/123456789/475793 |
| Appears in Collections: | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.