Discovering Travel Spatiotemporal Pattern Based on Sequential Events Similarity

Algorithm 5

Travel sequence recommendation algorithm (TSRA).

	Input: Historical Travel sequences set HS = {L11, L12, …, L1n} of user A, city a, historical travel sequences HSA = {L21, L22, …, L2m} of city a, historical travel sequences HSAU of all users
	Output: Travel recommendation sequences of city a for user A
(1)	newMedoids = TSCA (HSAU) //Cluster historical travel sequences of all users
(2)	for i ⟵ 1 to newMedoids.size do
(3)	for j ⟵ 1 to n do
(4)	userpre[i] ⟵ userpre[i] + TSCA (newMedoids[i], L1j);
(5)	userpre[i] = userpre[i]/n;
(6)	end for
(7)	end for
(8)	for t ⟵ 1 to m do
(9)	for r ⟵ 1 to newMedoids.size do
(10)	cityseq[t][r] = TSSA (newMedoids[r], L2t);
(11)	end for
(12)	end for
(13)	sim ⟵ 0;
(14)	for t ⟵ 1 to m do
(15)	if sim < CosSim (userpre, cityseq[t]) then
(16)	sim ⟵ CosSim (userpre, cityseq[t]);
(17)	outputseq ⟵ L2t;
(18)	end if
(19)	end for
(20)	return outputseq;