A model based on to cluster and predict the presence of malign breast cancer. The algorithms are applied to an open dataset of 569 patients that have either benign or malign breast cancer. First the article presents a detailed data description that is followed by a correlation matrix analysis and a regression analysis to verify the select the best variables for the clustering analysis and the predictive model. The clusterization is realized with the fuzzy c-Means algorithms. The prediction is performed by the Artificial Neural Network (ANN) Multilayer Perceptron (MLP) algorithm showing the best performance in comparison with the results of other machine learning algorithms. The use of all the tools in a definite sequence is the model proposed in this work providing an auto consistent approach to automatize the risk calculus with a good efficiency. The proposed approach can be applied for other typologies of cancers.