$8.$ Answer the following questions regarding classification tasks:

$($a$)$ How can a classification dataset be used to evaluate the capacity of a learning algorithm

to produce classifiers that generalise to unseen data?

$($b$)$ Consider the two types of errors $($false positives and false negatives$)$ in a binary classification task $($a classification problem with two classes$).$ Explain why one of these errors

may be worse than the other.

$($c$)$ Let f$1,$ f$2,$ and f$3$ be three different classifiers trained on the same training set. Suppose

that someone decides to use f$2$ for a specific application because it obtains the highest

accuracy on the training set. Explain why this methodology is flawed, and suggest a

better alternative.