Algorithm 2: Neighbor Query in DBSCAN
Input: a set of GPS trajectory $trj=\left\{p_1,p_2,\cdots ,p_m\right\}$, parameters ${\epsilon }_l,{\epsilon }_s$ Output: a set of $trjs=\left\{tr{j}_1,tr{j}_2,\cdots ,tr{j}_n\right\}$
1.	$trjs\leftarrow \varnothing$;
2.	$trj{s}_{nei}\leftarrow$ rtree.query($trj.mbr$);
3.	$curPaths\leftarrow trj.candidate$;
4.	foreach ($t in trjs\_nei$)
5.	$neiPaths\leftarrow$$t.candidate$;
6.	if ($dis(t.p_1,trj.p_1$)$<{\epsilon }_l and dis(t.p_n,trj.p_m$)$<{\epsilon }_l$)
7.	$d_n\leftarrow \mathrm{JaccardDissimilarity}\left(curPaths,neiPaths,{\epsilon }_s\right)$
8.	if ($d_n<{\epsilon }_s$)
9.	$trjs\leftarrow t$
10.	end if
11.	end if
12.	end foreach