Article

Article title REDUCTION OF DIMENSION STATE-SPACE PROBLEM OF ANALYSIS OF NETWORK TRAFFIC
Authors V.A. Nesterenko, A.A. Taran
Section SECTION III. METHODS AND MEANS OF NETWORK SAFETY
Month, Year 12, 2011 @en
Index UDC 519.254, 004.056
DOI
Abstract The article is devoted consideration of possibility of reduction of number of characteristics used at the analysis of a system. The problem of decrease in number of characteristics is very important by working out and creation of systems of intrusions detection: With increase in number of characteristics quality of systems of intrusions detection improves on the one hand and speed decreases on the other hand. Two methods are considered: Method of principal component analysis (PCA) and Fisher"s linear discriminant analysis (LDA). The estimation of efficiency of these methods and examples of their practical use is spent at the analysis of the network traffic.

Download PDF

Keywords Method of principal component analysis – PCA; Fisher's linear discriminant analysis – LDA; decrease in dimension of the data; detection of intrusions; the analysis of the network traffic.
References 1. RFC-1213. Management Information Base for Network Management of TCP/IP-based internets: MIB-II. Network working group, http://www.ietf.org/rfc/rfc1213.txt.
2. Данные, использованные на соревнованиях KDD CUP 99. http://sigkdd.org/
kddcup/index.php?section=1999&method=info.
3. Veksler О. Лекции по курсу распознавание образов. Университет Western Ontario, 2004. http://www.csd.uwo.ca/~olga/courses/CS434_541a/Lectures.pdf.
4. Богачев К.Ю. Практикум на ЭВМ. Методы решения линейных систем и нахождения собственных значений. – М.: МГУ им. Ломоносова, 1998.
5. Fisher R.A. The use of Multiple Mesurements in Taxonomic Problems. Annals of Eugenics. – 1936. – Vol. 7, part II.
6. Коллекция вредоносного трафика в формате pcap. https://www.openpacket.org/capture/by_category?category=Malicious.

Comments are closed.