Intelligent access control and intrusion detection techniques For web database security

Abstract

In this thesis an intelligent framework for effective access control newlineand Intrusion Detection has been proposed and implemented in order to newlineprovide effective Security for web databases The architecture of this system newlineconsists of two major subsystems namely Access Control Subsystem and newlineIntrusion Detection Subsystem The access control subsystem provides newlineeffective security through new SpatioTemporal Role Based Access Control newlineTechniques In addition it provides an additional access control feature based newlineon action status model The Intrusion Detection Subsystem provides newlinetechniques for effective classification of data This susbsystem has four major newlinemodules namely the preprocessing module outlier detection module newlineclassification module and decision making module newlineThe preprocessing module that consists of two different agents newlinenamely feature selection agent and feature validation cum ranking agent The newlinemain function of the feature selection agent is to determine a minimal feature newlinesubset by eliminating redundant and useless features from the dataset using a newlineFuzzy Rough Set Approach The feature validation and ranking agent collects newlineall the features selected by feature selection agent and checks whether they newlineare valid or not and ranks all the valid features using a Ranking Algorithm newline newline