Fit any four relevant supervised machine learning prediction models with the target variable hiv$\_$sero using Orange software. Give details of the methods you employed, features you selected and why, and sampling technique$/$s done.

Show and comment on the ROC curves and confusion matrices indicating what features those persons misclassified had different from the others. Then interpret one model and comment on how this performed compared to the model with the results as shown below.