A hybrid sampling combining Smote and RF algorithm for cancer chemotherapy protocols Classification

Breast Cancer (BC) is a network of cells that grow abnormally in the breast. If BC is not properly treated with the appropriate cancer chemotherapy protocols, it is at risk of causing death. This research aimed to combine Synthetic Minority Over-sampling (Smote) and Random Forest (RF) methods for BC chemotherapy protocols classification. Smote was used to balance the data, while RF was used to classify chemotherapy protocols data. The real data was produced by collecting medical and personal data from 601 patients with BC at the University Hospital Center (UHC) Mohammed VI of Marrakech in Morocco. Data of women diagnosed with BC from January 2018 at UHC were assessed. The results showed that the use of Smote for data augmentation can increase the performance of the RF classification method based on accuracy. There was an increase of 26% in accuracy. Time is an hyperparameter to be improved.