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

Table 2

Precisions of different algorithms (the best value of each row is shown as 60.52, the suboptimal value of each row is shown as 59.84, and unit is %).

N’	SVD	SVD++	NMF	Slope One	KNN	KNN + Zscore	KNN + means	KNN Baseline	Co-Clustering	BPR	Our_HS	Our_Mix
1	59.80	59.29	58.60	59.05	57.31	58.42	60.52	58.08	58.60	37.77	5.33	60.03
2	58.91	58.82	57.86	58.24	56.43	59.20	59.80	58.96	57.66	37.58	8.62	62.52
3	60.27	60.76	57.43	58.86	57.32	59.32	60.62	58.98	58.41	39.32	7.81	63.07
4	59.51	61.75	57.96	58.93	57.68	60.35	60.60	59.36	58.80	39.77	8.62	65.86
5	59.07	60.69	57.50	59.06	57.83	60.39	59.84	60.29	58.87	40.05	8.38	65.75
6	60.38	59.21	58.59	59.73	56.10	60.28	60.39	58.67	58.65	40.49	8.77	65.99
7	58.74	61.15	58.98	59.98	57.56	60.48	60.47	60.60	59.26	40.56	9.08	66.87
8	60.07	61.15	59.06	60.02	58.15	60.67	60.51	60.69	59.57	41.28	9.01	66.98
9	61.43	61.49	59.36	60.25	58.79	60.76	60.75	60.84	59.69	41.04	8.51	67.08
10	62.16	62.33	59.40	60.32	59.42	61.15	60.83	59.98	60.03	41.67	9.08	67.69
	60.04	60.66	58.47	59.44	57.66	60.10	60.43	59.65	58.95	39.95	8.32	65.18