We've demonstrated how ID$3($Iterative Dichotomiser $3)$ Algorithm works.

In Python implement your own code to demonstrate the instability using this dataset, by following the steps:

$($a$)$ Turn the dataset into a proper format, such as in arff used in Weka.

$($b$)$ Generate the decision tree using the ID$3$ algorithm.

$($c$)$ Remove a small number of examples, out of $x_1,$dots,$x_{12}$ so that the generated decision tree is different from the one in $($b$).$

$($d$)$ Add a small number of examples, different from $x_1,$dots,$x_{12}$ so that the generated decision tree is different from the one in $($b$).$

$($e$)$ Report on your own process and the discuss the findings.