Ecological Systems and Devices Annotation << Back Forecasting of Drinking Water Quality Physical and Chemical Indicators With the Application of Machine Learning Kuvayskova Yu.E., Klyachkin V.N. To reduce the time to eliminate dangerous situations, it is necessary to predict with suffi cient accuracy possible violations of the quality of drinking water. The initial data in the task under consideration are the results of monitoring the indicators of water coming from the water source, the dose of reagents (coagulants and fl occulants) used for water purifi cation, as well as the indicators of the quality of drinking water coming out of the treatment system for the previous period of time. It is necessary to predict the values of drinking water quality indicators. Standard regression analysis methods often do not provide the necessary accuracy. In this case, you can use machine learning methods: the article discusses the use of the support vector machine, random forest and gradient boosting. At the same time, it is important to select the optimal values of the hyperparameters of these methods, which signifi cantly affect the accuracy of forecasting. The calculation technology is shown by the example of estimating the aluminum content in drinking water after its purifi cation. Keywords: Water Treatment, support vector method, decision tree busting, random forest, hyperparameters. DOI: 10.25791/esip.2.2024.1428 Pp. 03-08.