Non-response in surveys is a source of error in survey results, and national statistical organizations are always looking for solutions to control and reduce it. Predicting non-response sampling units in surveys before conducting the survey is one of the solutions that can greatly help reduce and eliminate the problem of non-response in surveys. With recent technological developments and the facilitation of complex calculations, it has become possible to use statistical learning methods, such as regression and classification trees or support vector machines, in many problems, including predicting non-response of sampling units in surveys. In this article, while reviewing the above methods, non-response sampling units in a field survey are predicted using them, and it is shown that the combination of the above methods is more accurate in correctly predicting non-response than any of the individual methods.