Ensemble Approach Based on Bagging and Boosting for Identification the Computer System State

Abstract

The efficiency of using machine learning technology to detect the state of a computer system has been studied. A set of different classifiers and ensembles of classifiers was developed, their training, cross-checking (testing) on real data were carried out. Based on the research results, two methods for identifying the state of a computer system are proposed using an ensemble of decision trees based on boosting and bagging as a classifier. These classifiers were modified due to a special procedure for selecting the optimal parameters for the functioning of the classifiers as well as through the use of the initial data preprocessing procedure. The developed methods are implemented in the software and are investigated in solving the problem of identifying the state of the computer system. The efficiency of the developed classifiers has been evaluated. Prospects for further research may be the development of an ensemble of fuzzy decision trees based on the proposed methods, optimization of their software implementation.