Artificial Neural Networks are a branch of artificial intelligence whose way of working mimics or imitates the human brain. In Artificial Neural Networks, there is a Backpropagation method that can be used to solve various problems, especially in the field of prediction. Backpropagation Neural Network method is a method that can be used to predict student graduation. Backpropagation is a supervised learning algorithm. There are two stages used in the Backpropagation method, namely the training stage and the testing phase. The data used were 200 samples of academic graduation data from students of the Informatics Engineering Study Program, Muhammadiyah University of Purwokerto in 2012-2015. The variables used for the prediction are 6 variables, namely the Achievement Index (IP) semester 1 to 4 and the number of semester credit units (SKS) 3 to 4. This prediction is carried out using input neurons 6, output neurons 1 and 12 neurons in a hidden layer. From the calculation results, the prediction accuracy percentage is 92%.