Ecological Systems and Devices Annotation << Back Evaluation of Carbon Monoxide Content in Flue Gases During the Operation of Burner Devices Generalov D.A., Kovalnogov V.N., Klyachkin V.N., Sherkunov V.V. Reducing the level of carbon monoxide involves assessing and predicting its content in fl ue gases, depending on the performance of the burner devices. Based on the known results of operation and testing of the burner device, the values of the performance indicators and the corresponding values of the carbon monoxide content are recorded. It is necessary to build a mathematical model that links the response – the content of carbon monoxide – with the factors that affect its level. The aim of the study is to develop an algorithm for building such a model that provides the necessary accuracy in estimating the carbon monoxide content for the given values of the burner device performance indicators, based on machine learning methods with optimization of the hyperparameters of the corresponding models. Keywords: regression model, machine learning, decision tree busting, random forest, support vector method, hyperparameters. DOI: 10.25791/esip.2.2024.1433 Pp. 51-56.