Wear resistance testing is essential when designing parts intended to operate in environments that expose the parts to high wear. Wear tests are expensive and time-consuming, and require special test equipment. The use of machine learning algorithms to predict the amount of wear is a potentially effective means of eliminating the disadvantages of experimental methods, such as cost, labor, and time. In this study, data on the wear loss of epoxyfuran coatings were experimentally obtained. The mechanical properties of the coating were determined by measuring the coefficient of friction, linear wear rate, and temperature in the friction zone during sliding. Wear tests were performed at different loads (from 0 to 3 MPa) at a relative sliding speed of 0.5 m/s. In this study, the models were developed using machine learning algorithms (artificial neural network), based on the data set obtained from the wear experiments. The error value was calculated as 0.9729 for the models.