DEVELOPMENT OF MODELS AND ALGORITHMS FOR EVALUATING THE FUNCTIONING OF INTRUSION DETECTION SYSTEMS

The relevance of the article is due to the need to improve existing intrusion detection systems in the context of a constantly changing arsenal of tools and techniques of intruders. Classic intrusion detection systems algorithms based on signature and behavioral analysis do not provide a sufficient degree of network security and cannot prevent dynamic attacks on systems. The development of new algorithms and models will improve the overall security of the network structure, reduce the number of false positives and minimize damage from computer attacks. Artificial immune systems use approaches to combat malicious influence similar to the mechanisms observed in living organisms. Namely, the detection of viruses and the development of an immune response – antibodies. This approach allows computer systems to further learn during operation, independently identifying computer viruses by their activity and independently developing means of combating malicious code.