Academic failure among first-year university students has long raised a large number of arguing. Many educational psychologists and researchers have tried to understand and then explain it. The target of this research is to adapt a Neural Network Approach that can be used to predict student's success-failure risk. Multilayer feed forward back error propagation artificial neural network model has been adapted and trained with available data driven from existing academic acceptance system results. A performance analysis is done to analyze the effectiveness of such model for this particular problem. The result of this study shows that the neural network model can predict students' performance, if the current and future samples have similar characteristics.