Multiview Ensemble Method for Detecting Shilling Attacks in Collaborative Recommender Systems

Table 10

Comparison of running time for seven detection methods.

Algorithm	Training time (s)			Detection time (s)
	Netflix dataset	Amazon review dataset		Netflix dataset	Amazon review dataset
		random sampling method	group sampling method		random sampling method	group sampling method
MV-EDM	74.59	42.81	75.26	6.38	15.92	7.28
MEL-once	29.58	18.22	44.53	5.72	9.25	6.36
MV-SVM	20.79	12.73	36.28	1.36	7.46	5.10
SVM-TIA	487.58	196.38	529.65	18.75	59.37	32.71
RAdaBoost	25.07	24.19	37.58	2.77	27.15	12.67
RF-13	632.11	304.86	712.68	26.42	90.27	45.83
DWT-SVM	115.31	64.65	104.75	15.44	72.38	28.50