Algorithm 1 One hierarchical clustering tree building
Input: D: features dataset; N: size of D; K: branching factor; FL: maximum leaf size
Output: T: clustering tree
1:	generate head node of the tree T, represented with header
2:	if N < FL then
3:	expand all points in D as child nodes of header
4:	else
5:	we choose K points {q1, q2, ⋯ , qK} randomly from D, K clusters C = {C1, C2, ⋯ , CK}, qk is the center of Ck
6:	for i = 1 → N do
7:	min_length = MAX_NUMBER, k = 0
8:	for j = 1 → K do
9:	if $\widetilde{{\text{p}}_i{\text{q}}_j}\le \mathit{min}\_\mathit{\text{length}}$ then
10:	$\mathit{min}\_\mathit{\text{length}}=\widetilde{{\text{p}}_i{\text{q}}_j}$, k = j
11:	end if
12:	end for
13:	end for
14:	for i = 1 → K do
15:	recursively apply the algorithm to the points in Ci
16:	end for
17:	end if