Download the following datasets: circles $0\mathrm{.}3,$ moons$1,$ spiral$1,$ twogaussians $42$ and

halfkernel.

Run the following classifiers on all downloaded datasets and apply $10-$fold cross validation:

a$.$ LDA $($linear discriminant analysis$)$

b$.$ Quadratic $($quadratic discriminant analysis$)$

c$.$ Na$$ve Bayes: Choose a distribution and explain why you chose it$.$

d$.$ SVM: Choose a kernel and explain why you chose it$.$

For each classifier, provide the following five performance measures: PPV$,$ NPV$,$

specificity, sensitivity and accuracy.

Plot the samples of all datasets $($in separate plots$)$ as points in the $2D$ space, using a different

color and point shape for each class.

Discuss the performance $($accuracy only$)$ for each classifier and dataset individually: Why

do you think a particular classifier on a dataset is good$/$poor$?$ Give very strong reasons,

including mathematical formulas and$/$or arguments, if needed.

In one paragraph, discuss if you have used any form of help by using Generative AI tools,

and how your text is different$/$enhanced from that automatically generated text.