Breast Cancer is the common cancer among the women in the world.  The aim of this paper therefore, is not to inhabit on the prediction techniques themselves, but rather concentrate on comparison of results produced from these two alternatives (Neural Networks and Multiple Regression), yet complementary techniques.  In order to gauge the success (or otherwise) of either techniques, a comparative analysis of prediction performance must be made [1]. The models use the conventional statistical technique multiple regression and artificial neural networks.   Performance analyses using mean percentage error, mean absolute percentage error and percentage of correctness (generalisation). Results reveals that ANNs model perform well, having low mean absolute percentage error values indicating that predictor variables were reliable inputs for modelling breast cancer database.  Overall, the neural network model performs slightly better as it was able to predict up to 97.14 % generalisation compare with multiple regression just only 82.10 % generalisation