IMPROVING TRAINING OF NEURAL NETWORK INTRUSION DETECTION SYSTEMS USING FUZZY LOGIC
Loye RayA AzarS El-SaidY BodyanskiyO VynokurovaG SetlakD PeleshkoP MulesaL DahanabalS ShantharajahZ HaoZ ZhangH ChaoM HassanX JingY BiH DengM MasdariH KhezriM Motahari-NezhadM MazidiM MukoseraT MpofuB MasaitiM NemissiH SeridiH AkdagK QaddoumE HinesD IliescuQ QassimA PatelA Mohd-ZinV OjhaA AbrahamV SnaselL RayL RayH FelchL RayH ShahparastE MansooriK ShihabudheenG PillaiA ShubairS RamadassA AltyebP SouzaT RezendeA GuimaraesV AraujoL BatistaG SilvaV AraujoM VarnamkhastiN Hassan
Abstract
Today's threats to networks use various techniques that attempt to penetrate protective barriers.This taxes current intrusion detection systems to stay up with these attacks.Training a neural network intrusion detection system is important to detecting dynamic threats facing a network.However, keeping them trained in such a dynamic threat environment can prove challenging.Therefore, finding a fast method of training an IDS is important.This paper shows how the use of a fuzzy logic inference system can improve the training time for neural network intrusion detection systems.Using a combination of both fuzzy inference system and neural network techniques proved successful in reducing the convergence time of intrusion detection systems.
Metrics
Topics
Related Documents
Network Traffic Intrusion Detection System Using Fuzzy Logic and Neural Network
Mrudul DixitRajashwini Ukarande
Hybrid Intrusion Detection Systems (HIDS) using Fuzzy Logic
Bharanidharan ShanmugamNorbik Bashah
Improving Intrusion Detection Accuracy using Convolution Neural Network
Vara, Samuel SudheerAshalata PanigrahiManas Ranjan Patra
Improving Intrusion Detection Accuracy using Convolution Neural Network
Vara, Samuel SudheerAshalata PanigrahiManas Ranjan Patra