Novel Recommendation System for Tourist Spots Based on Hierarchical Sampling Statistics and SVD++

Table 3

Recall values of different algorithms (the best value of each row is shown as 16.13, the suboptimal value of each row is shown as 17.89, and unit is %).

N’	SVD	SVD++	NMF	Slope One	KNN	KNN + Zscore	KNN + means	KNN Baseline	Co-Clustering	BPR	Our_HS	Our_Mix
1	13.31	15.94	14.92	13.39	15.93	15.36	15.04	14.57	16.13	9.38	0.33	15.94
2	15.36	17.89	15.55	15.40	16.41	14.08	14.36	14.71	16.73	10.25	0.80	17.96
3	13.40	17.93	12.65	13.95	15.87	13.57	14.39	15.23	17.63	9.91	1.12	17.83
4	14.81	16.54	14.30	14.01	16.27	15.48	14.61	15.63	16.77	10.42	1.80	17.84
5	14.99	18.29	15.76	14.88	16.32	15.51	14.86	15.42	17.16	10.62	2.11	18.69
6	16.21	14.65	15.78	15.80	16.61	16.62	14.97	16.31	17.47	9.89	2.34	17.66
7	17.97	18.86	16.13	16.39	16.85	16.73	16.54	16.47	18.85	11.64	3.14	19.20
8	17.20	18.65	16.38	16.45	16.94	16.74	16.81	16.58	16.13	12.08	3.43	19.68
9	17.29	18.75	16.83	17.17	16.96	16.82	16.23	16.74	17.49	13.07	4.81	20.20
10	18.16	18.25	16.28	17.39	17.56	16.70	16.92	17.08	17.64	13.12	5.56	21.70
	15.87	17.57	15.46	15.48	16.57	15.76	15.47	15.88	17.20	11.04	2.54	18.67