Forecasting the enrollments of new students in bachelor's systems became an urgent desire in the majority of higher education institutions. It represents an important stage in the process of making strategic decisions for new course's accreditation and optimization of resources. To gain a deep view of the educational forecasting context, the most used machine learning and statistical approaches are discussed and analyzed. These methods were applied over student data collected from the enrollment of new students in the faculty of literature and Human sciences between 2003 and 2019. The main result of this study is the development of a forecasting model that provides the most accurate values with a minimum of errors.l