Abstract
The article deals with the actual task of identifying abnormal situations arising in information and telecommunication systems (ITCS) using artificial intelligence methods. Analysis of the processes of functioning of ITCS devices and nodes in real conditions shows that over time the properties of the analyzed characteristics can be transformed. Various factors cause the ranges of the analyzed values of the target variables to shift. The appearance of external and internal events at various points in time changes the states of both individual telecommunication devices and the system as a whole. It is necessary to analyze a large number of heterogeneous qualitative and quantitative characteristics of traffic and processes of functioning of telecommunication devices. The proposed method of monitoring the state of ITCS is based on taking into account the factors influencing the change in the properties of target variables, which, in turn, allows segmentation of the data sample. The novelty of the proposed solution lies in the fact that a classification algorithm with the best quality indicators is used for individual segments of the processed data. Taking into account the current and expected situations, it is possible to form samples based on the identified impacts. On the basis of the NSL-KDD dataset, segmentation of the data set is carried out, taking into account the influence of factors on the ranges of values. The processing of factors using the search function of the time series layout, which was used to segment the data set, is shown. As a result of data segmentation, a finite number of non-overlapping measurable subsets of data were obtained, each of which was fed to the classification algorithm. The evaluation of the quality of the classification algorithms was carried out. Based on the results obtained, it is possible to build various classification models of the ITCS information security (IS) status. The proposed approach makes it possible to improve the quality indicators of classification algorithms in dynamically changing conditions.