Predicting The Recurrence of Colorectal Cancer in Patients Using Ensemble Methods

Among the top ten death-causing diseases in the world, cancer isconsidered to be one. There are about 100 different forms of cancer, and cancer prediction is critical to the advancement of data mining applications. The development of adequate methods for locating cancer instances will go a long way toward improving cancer patient therapy diagnosis. The term “cancer” refers to a group of disorders. All malignancies are caused by cells that continue to divide without stopping and spread to surrounding tissues. The ensemble model can be effectively used to prognosticate the recurrence of colorectal cancer in clinical patients. According to current data from Globacan India, over 27,000 cases were documented in the preceding year, with about 20,000 people dying. In this project, ensemble approaches are utilized to predict cancer recurrence. Basic classifier approaches such as SVM, and RF, as wellf as ensemble classifiers of bagging, were employed to construct classier models. According to the findings, the constructed models’ prediction performance was normally good. In terms of accuracy, ensemble models outperform when compared to Simple classifiers. In a model comparison, the ensemble voting technique outperformed all other models by a significant margin.