Misuse Detection System Using Artificial Neural Network

The system developed for Projek Sarjana Muda (PSM) is entitled Misuse Detection System using Artificial Neural Network. It is a system that detects misuse based on packet with the help of neural network and using feedforward backpropagation technique. The system operates offline and use data from DA...

Full description

Saved in:
Bibliographic Details
Main Author: Amirah Nabihah , Jahidin
Format: Monograph
Published: UTeM 2008
Subjects:
Online Access:http://library.utem.edu.my:8000/elmu/index.jsp?module=webopac-d&action=fullDisplayRetriever.jsp&szMaterialNo=0000046992
http://library.utem.edu.my:8000/elmu/index.jsp?module=webopac-d&action=fullDisplayRetriever.jsp&szMaterialNo=0000046992
http://eprints.utem.edu.my/2044/1/Misuse_Detection_System_Using_Artificial_Neural_Network_Amirah_Nabihah_Bt_Jahidin_QA76.87.A44_2008_-24_pages.pdf
http://eprints.utem.edu.my/2044/2/Misuse_Detection_System_Using_Artificial_Neural_Network_Amirah_Nabihah_Bt_Jahidin_QA76.87.A44_2008_..pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The system developed for Projek Sarjana Muda (PSM) is entitled Misuse Detection System using Artificial Neural Network. It is a system that detects misuse based on packet with the help of neural network and using feedforward backpropagation technique. The system operates offline and use data from DARPA during training and testing. The system believes capable to identify network attacks by training the neural network to recognize data that contain normal packet against data that contain misuse packet that probably intent to attack the network. The project is targeted to deliver a system that able to identify normal and misuse packet. Most current approaches to misuse detection involve the use of rule-based expert systems to identify indications of known attacks. However, these techniques are less successful in identifying attacks which vary from expected pattems. Artificial neural networks provide the potential to identify and classify network activity based on limited, incomplete, and nonlinear data sources. This paper presents an approach to the process of misuse detection that utilizes the analytical strengths of neural networks which at the end of the project, Misuse Detection System using Artificial Neural Network is proved capable to detect misuse in network by identified misuse and normal packet that results with 98% accuracy.