Generation of Network Behaviour Descriptions Using MCA Based on TAM

Tan, Zhiyuan

Abstract

In this paper, a multivariate correlation analysis technique based on triangle area map is introduced. The technique is applied for network traffic characterization and provides quality network behaviour descriptors for intrusion detectors to use. The effectiveness of the multivariate correlation analysis technique is evaluated using a misuse detector and an anomaly detector. The evaluation is conducted on the KDD Cup 99 dataset. The results show that the multivariate correlation analysis technique supplies with high quality network behaviour descriptors, which facilitate both the detectors with encouraging performance in detecting DoS attack network traffic.

Citation

Tan, Z. (2013, March). Generation of Network Behaviour Descriptions Using MCA Based on TAM. Paper presented at Kaspersky Lab's 2013 Annual Student Cyber Security Conference

• Presentation Conference Type: Conference Paper (unpublished)
• Conference Name: Kaspersky Lab's 2013 Annual Student Cyber Security Conference
• Start Date: Mar 21, 2013
• End Date: Mar 23, 2013
• Deposit Date: Jun 16, 2017
• Keywords: multivariate correlation analysis, anomaly-based detection, denial-of-service attack, misuse-based detection
• Public URL: http://researchrepository.napier.ac.uk/Output/948354