Hi there, could you consider writing these problems out in text/markdown so that I can type this instead of simply copy/pasting images?
Will solutions be posted as well? thanks!
Normally they would, but I guess this is a curious exception.
But in the meantime, here are my solutions:
Part 1.1:
Supervised learning uses labeled data, so it would be \boxed{\text{C}}, which has no labels beforehand. D and B here would have 2 labels, positive or negative, so they would be out.
Part 1.2 and 1.3:
Here is a link to a similar question from the sample, and the solution.
Part 1.4:
Here is a link to a similar question from the sample, and the solution.
An F-1 score is a mix of precision and recall, and is only high when both of them are high, so it is ideal. Accuracy is different from those 3, and only measures overall correctness. You should use F1 when the dataset is imbalanced, because all the other cases are irrelevant.
Part 1.5:
The answer is (B.), this one you kinda have to understand UMAP to know.