Friedman statistical analysis and cosine threshold method for TCP based malware detection

Mohammad, Nur Ameera Natasha (2017) Friedman statistical analysis and cosine threshold method for TCP based malware detection. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

	Text (24 Pages) Friedman statistical analysis and cosine threshold method for TCP based malware detection.pdf - Submitted Version Download (438kB)
	Text (Full text) Friedman statistical analysis and cosine threshold method for TCP based malware detection.pdf - Submitted Version Restricted to Registered users only Download (1MB)

Abstract

Intrusion Detection System (IDS) is a network security technology which inspects all inbound and outbound on computer network traffic and design for detecting suspicious patterns that attempts to perform security policy violation. IDS approach the goal to detect threats in various ways. Most of the IDS implement signature based which means that they operate almost as same as a virus scanner, by search for a known identity or any signature for each specific intrusion event. The Friedman test is used to test for differences between groups when the dependent variable being measured is ordinal and for continuous data that has violated the assumptions necessary to run the one-way Anova with repeated measures. As what has been described above, Friedman test has meaning of the non-parametric alternative to the one-way Anova with repeated measures. The problem statements for this project are that the lack of approach to examine the degree of behaviour of each packet more accurate and to ensure whether the unforeseen packets behaviours contain anomalous and non-anomalous activity is hard to differentiate. For the objective, it is to distinguish the degree of packet behaviour using Friedman statistical base analysis for detecting behaviour more correctly and also to differentiate the anomalous and non-anomalous packets behaviour more accurately using scoring method. For methodology, this project represents the method in Friedman statistical analysis for detecting packet behaviour including the specific steps that will undertake to produce accurate outputs. There are few steps to be focus in this project which include data preparation, data scoring where include anomaly score and normal score and last step is analyse data. Analyse data divided into standard deviation, mean method and normal. This project will contribute on proposing the new technique to identify intrusion by classifying activity as either anomalous or normal. Other than that, this project also contribute in distinguish the degree of packet behaviour using Friedman statistical base analysis for detecting behaviour.

Actions (login required)

View Item