Identification of the Computer System State Based on Multidimensional Discriminant Analysis

Abstract

This paper describes the process of identifying the state of a computer system based on tracking changes in the file system under the condition that there is no reliable information about the state of objects in the external environment and the state of the system itself. In this case, the performance indicators of a computer system are random variables with unknown distribution laws. A method for identifying the state of the computer system based on tracking changes in the file system using classical discriminative analysis, taking into account the uncertainty of the input data, is proposed. For this purpose, a special method was developed for solving a fuzzy system of linear algebraic equations. The software was developed for tracking changes in the computer file system: creating, modifying, renaming, and deleting files and directories. The program also performs state classification based on a fuzzy discriminant classifier.