BROWSE
Detail
Conference Paper
An Efficient Clustering Algorithm for Classification of Computer Network Attacks
Cited - time in 
Share 
Share
- Authors
- Chi Yoon Jeong, Bum Hwan Chang, Jung Chan Na
- Issue Date
- 2007-07
- Citation
- International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC) 2007, pp.1356-1357
- Publisher
- IEEE
- Language
- English
- Type
- Conference Paper
- Abstract
- In this paper, we proposed an efficient clustering algorithm for classification of computer network attacks. The most existing visualization approaches display the traffic volume between two hosts or current status of network. However, proposed method analyzes the network traffic and display the group of network attack which has similar property using the clustering algorithm. The proposed method aggregates the traffic information using two elements from the five-tupules of Netflow data, and then calculates the dispersion degree about the remainder elements except protocol. At this time, traffic data having the similar dispersion degree have the similar characteristic of network attack. The proposed method performs the clustering in 2D by using two dispersion degree, port information, and the protocol information. We compared our algorithm to two conventional clustering algorithms: K-Means and SOMs. The experimental results show that the proposed clustering algorithm speeds up processing time.
- KSP Keywords
- Clustering algorithm, Computer network, Current status, Dispersion degree, Netflow data, Network Attacks, Network Traffic, Similar property, Traffic volume, k-Means, processing time