Open AccessVISUALIZATION AND PROCESSING OF INFORMATION SECURITY EVENTS BASED ON CICIDS DATA 17
Author(s) -
I.U. Zelichenok
Publication year - 2021
Publication title -
informacionnye tehnologii i telekommunikacii
Language(s) - English
Resource type - Journals
ISSN - 2307-1303
DOI - 10.31854/2307-1303-2021-9-4-49-55
Subject(s) - computer science , preprocessor , intrusion detection system , visualization , data pre processing , data mining , set (abstract data type) , feature extraction , information extraction , raw data , information visualization , machine learning , intrusion , information security , data set , feature (linguistics) , artificial intelligence , information retrieval , computer security , geochemistry , programming language , geology , linguistics , philosophy
At present, attacks on computer networks continue to develop at a speed that outstrips the ability of information security specialists to create new attack signatures. This article illustrates an approach to preprocessing raw data and visualizing information security events in a live dataset. It is shown how preprocessing and primary knowledge extraction for further use of the processed dataset in machine learning models can be used in the design of machine learning models for intrusion detection systems. A distinctive feature of the work is that the most relevant set CICIDS17 was taken as the studied dataset. Although traditionally considered popular such kits as DARPA2000 and KDD-99, which are more than 20 years old. The article also describes the criteria and characteristics that the set has.