$)$ For this exercise I want you to come up with a machine learning domain and answer some questions about it$.$ I

encourage you to choose a domain that interests you.

$$ Describe the domain of interest $($one or two paragraphs$).(3$ points$)$

$$ Describe $3$ to $6$ features that you think would be useful in this problem. Explain why. $(3$ points$)$

$$ Describe $2$ features that you think would not be useful in this problem. Explain why. $(3$ points$)$

$$ Use $2$ of the useful features to plot $6$ instances $(3$ of each class$)$ in $2$d space. Would a linear classifier work well

in this $(2$d$)$ problem $($explain why$)?$ Would nearest neighbor work well in this $(2$d$)$ problem $($explain why$)?(6$

points$)$

$($Very$)$ Rough and incomplete example of the kind of work I expect.

My domain is dog classification. In particular I want to build a classifier to discriminate between two classes.

$$ Class A: Hounds.

$$ Class B: Terriers.

For background I will describe each type of dog.

Hounds were traditionally bred to help hunters track and course game. There are two divisions of this

group: the sight hounds that hunt by sight and are speedy coursers with incredible flash and style, and the

scent hounds that hunt by smell alone.

Terriers originated in the Irish and British Isles $($except the Miniature Schnauzer$).$ They are intelligent,

lively dogs that are bred to hunt game and vermin.

One feature that I believe would be very useful in discriminate between the two classes is height $($note$,$ for dogs, the

height is measured at the shoulder$).$ This feature would be useful because most hounds tend to be tall $($although there

are exceptions, like the Dachshund$)$ and most terriers tend to be short.

For example, here are the $$ideal$$ heights for some dogs, as defined by the American Kennel Club.

Whippet $($hound$)$: $19$ to $22$ inches.

Pharaoh Hound: $22$ to $25$ inches.

Irish Wolfhound: $32$ to $34$ inches.

As we can see from the examples above....

Another feature.....

To the right are $6$ instances projected into $2$D space.

The X axis is $...$ and the Y axis is $...$ measured in