Algorithm 1. Pseudo-code for the k-NN Algorithm.
Input:
	$X$: training data
	$Y$: class labels of $X$
	$x$: unknown sample
	$S$: instance set
	$k$: number of nearest neighbor
Output: predicted class label
1:	//train phase
2:	for $i$ = 1 to $length\left(X\right)$ do
3:		Store training instance $X\left[i\right]$, class label $Y\left[i\right]$ into instance set $S\left[i\right]$
4:	end for
5:
6:	//inference phase
7:	for $i$ = 1 to $length\left(X\right)$ do
8:		Calculate distance between the unknown sample and the stored training instance $d\left(X\left[i\right], x\right)$
9:		Store calculated distance values into $D$
10:	end for
11:
12:	Find $k$ minimum distance values
13:	Vote which class is the most among $k$ minimum distance values
14:
15:	Return the class that has been voted by the majority