Survival Analysis Versus Machine Learning

The wpbc dataset (available at https://archive.ics.uci.edu/ml/machine learningdatabases/breast-cancerwisconsin/wpbc.data). It is available at https://archive.ics.uci.edu/ml/machinelearning databases/breastcancer-wisconsin/wpbc.names. We want to predict the probability of relapse (“recurrent”) at 24 months. To do this, you will compare the methods of survival analysis (Cox models, survival random forests,. . . ) with 1the classification methods. Performance measurements (including AUC) will be made on a test sub-sample consisting of 20 to 30% of the data (be careful to stratify well!).