Given a binary classification dataset, say having Y $=1$ or $0$ as the target variable and

X

$1$

and

X

$2$

as two numeric attributes, what does it mean to say that the dataset is linearly separable?

X

$1$

and

X

$2$

lie on a straight line

a

X

$1$

$+$

b

X

$2$

$+$

c

$=$

$0$

A straight line

a

X

$1$

$+$

b

X

$2$

$+$

c

$=$

$0$

can separate the points Y$=0$ from Y$=1$

Both $($Y$,$ X$1)$ and $($Y$,$ X$2)$ lie on a straight line

X$1,$ X$2$ and Y lie on a plane in $3-$D