Determination of oxidant exposure values is crucial in order to calculate the extent of oxidation of various particles in the target water such as micropollutants during the ozonation process. While it is somewhat possible to determine ozone exposure using in-situ sensors, realistic hindrances make it difficult to accurately calculate hydroxyl radical exposure values in large scale water treatment plants. To overcome this problem, prediction of oxidant exposures using modeling is a viable option. In this study, ozone exposure values themselves were used as potential input parameters alongside basic water characteristic parameters and ozonation experiment conditions. A user-defined response surface methodology (RSM) approach as well as various machine learning methods were utilized to develop models for different input parameter combinations to predict oxidant exposure values. Model training and validation results showed that using both initial ozone dose and ozone exposure value showed the highest prediction accuracy for hydroxyl radical exposure values.