Consider the College data set from ISLR's webpage: https://www.statlearning.com/resources-first-edition. Note that the data set can be accessed by clicking on (College.csv) link, and its explanation can be found via clicking in page 5 here: https://cran.r-project.org/web/packages/ISLR/ISLR.pdf, among other places

Build a predictive model for Grad.Rate or graduation rate of the colleges in the data set. Have in mind that the ultimate goal here is prediction, not inference. So, we would like to be able to have model that can, with relatively high accuracy, predict the outcome. Finally, demonstrate the robustness of that predictive engine via apply it to randomly selected and roughly 5% of the responses in your data (validation or test set). Choose whatever technique that you see fit.