Abstract
Breast cancer diagnosis and prognosis are two medical applications that pose a great challenge to the research community. Breast cancer has become the leading cause of death for women in the world. Recurrence of cancer is one of the biggest fears in the life of a cancer patient and thus one of the issues that affect their quality of life. The aim of this research is to improve the prediction of breast cancer recurrence using an ensemble learning technique and to provide a website that enables physicians to enter features related to a breast cancer patient and get the probability of breast cancer recurrence. This can help physicians make better treatment decisions for patients to increase their survival rate. All experiments are done in R-studio statistical programming suit, and its relevant packages. The results have shown that the best performing models are random forest with accuracy 0.6522, sensitivity 0.6250 and specificity 0.6593, then decision tree with accuracy 0.6261, sensitivity 0.63636 and specificity 0.62500, finally naive Bayes with accuracy 0.5913, sensitivity 0.4889 and specificity 0.6571.