Algorithm 1 Sampling-based Candidate Pruning
Input:
	Candidate k-Sequences Ck; Sample Database dbk; Privacy Budget ε4;
	Threshold θ; Maximal Length Constraint lmax;
Output:
	Potentially Frequent k-Sequences PF;
1:	${\mathit{\text{db}}}_k^{\prime }$ ← Transform sample database dbk; \\ see Sec. V-A
2:	θ′ ←- Relax threshold θ; \\ see Sec. V-B
3:	PF ← discover_potentially_frequent_sequences (Ck, ${\mathit{\text{db}}}_k^{\prime }$, ε4, θ′);
4:	return PF;