if a and b then x
if c and d then x
if e and f then x
if g or h then x
Consider the data here. Randomly split it so that
4/5 as training and 1/5 as testing.
Implement k nearest neighbor methods and calculate
the testing accuracy of k = 1, 3, 5.
Note that in this data set, the end of each line is the
class label.