In R, using the data in wine.csv, use one of the three supervised classification algorithms (logistic regression and knn,..) determine the class of wine based on 13 variables. Use an 80/20 split to test your model.

Then use the data in wine.csv again, this time ignore the class and run a clustering algorithm to see if you can find the 3 clusters, Does the clustering algorithm think 3 is the correct number of clusters?

Dataset:

https://www.kaggle.com/datasets/hrdkcodes/wine-data