Machine Learning in Intrusion Detection

دانشگاه آزاد اسلامی اصفهان(خوراسگان)

منو

رکورد قبلی

رکورد بعدی

" Machine Learning in Intrusion Detection "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	149779
Doc. No	:	ET21571
Main Entry	:	YIHUA LIAO
Title Proper	:	Machine Learning in Intrusion Detection
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	Three general approaches to intrusion detection . . . . . . . . . . . . . . . .Pseudo code for the WN classifier algorithm for anomaly detection . . . . .Performance of the kNN classifier method expressed in ROC curves for the tfeidf weighting method . False positive rate vs attack detection rate for k 5 .10and25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ROC curves for tf . idf weighting ( k 1 0 ) and frequency weighting ( k 1 5 ) . .(a) A simplified diagram of an evolving connectionist system for unsupervisedlearning . The system has n input nodes and m pattern nodes . There is aconnection from each input node to every pattern node . Some connectionsare not shown in the figure . (b) An evolving connectionist system that has anadditional fuzzy input layer . The task of the fuzzy input nodes is to transferthe input values into membership degrees . . . . . . . . . . . . . . . . . . . .Pseudo code for adaptive anomaly detection . . . . . . . . . . . . . . . . . .Flow graph representation of the Fuzzy ART algorithm . . . . . . . . . . . .A two-dimensional sphere (solid line) containing most of the data . Enclosedin a circle are four support vectors on the boundary . . . . . . . . . . . . . .Experimental results for various v values on the artificial data . (a) Compari-son of error estimate. 10-fold cross validation error and false positive rate onthe testing set . (b) True positive rate vs . v . . . . . . . . . . . . . . . . . . .Comparison of error estimate. leave-one-out error and false positive rate ontesting data . (a) User 9; (b) User 24; (c) User 42; (d) User 43 . . . . . . . .Experimental results for various 7 values on the KDD data (v = 0.001). (a)Comparison of error estimate and 5-fold cross validation error on the trainingset . (b) False positive rate and false negative rate on the testing set . . . .Learning curves for masquerade data . (a) User 5; (b) User 6 . . . . . . . . .Extensive form for Matching Pennies . . . . . . . . . . . . . . . . . . . . . .Normal form of Matching Pennies . . . . . . . . . . . . . . . . . . . . . . . .Game example . . . . . . . . . . . . . . . . . . . .tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL44745.pdf
Title and statement of responsibility and	:	Machine Learning in Intrusion Detection [Thesis]