Based on the World Health Organization (WHO), breast cancer ranks eighth which causes the largest mortality rate in the world. Based on the data from the Ministry of Health of the Republic of Indonesia, breast cancer is at second position after cervical cancer. In Indonesia, more than 80% of cases were found difficult to make treatment efforts because the cases are at an advanced stage.Prediction of breast cancer is used artificial neural networks backpropagation method by dividing data into two parts, training 70% while testing 30% and for optimizing parameters such as the number of hidden neurons and learning rate to achieve accurate results. This study aims to provide an accurate diagnosis of breast cancer. The method for evaluating the accuracy of Backpropagation is the Confusion Matrix.From the results of evaluating the accuracy of Backpropagation with Confusion Matrix shows that Backpropagation Neural Network produces a testing value of 94.634% and training of 99.372%.