Collaborative QoS Prediction for Mobile Service with Data Filtering and SlopeOne Model

Table 4

Accuracy comparison (a smaller value means higher accuracy).

Model	Training set density (TD) — response time
	MAE	NMAE	MAE	NMAE	MAE	NMAE	MAE	NMAE
	UserMean	0.8829	1.0879	0.8766	1.0861	0.8785	1.0861	0.8749	1.0849
ItemMean	0.7318	0.9019	0.7121	0.8823	0.7101	0.8779	0.7031	0.8718
SlopeOne	0.7126	0.8757	0.6929	0.8560	0.6886	0.8544	0.6843	0.8492
IPCC	0.6833	0.8422	0.6248	0.7741	0.6037	0.7464	0.5849	0.7253
UPCC	0.6763	0.8335	0.6304	0.7811	0.6172	0.7630	0.6053	0.7506
WSRec	0.6542	0.8063	0.6157	0.7628	0.6026	0.7450	0.5825	0.7223
MF	0.6441	0.7939	0.5405	0.6697	0.5207	0.6438	0.5023	0.6229
FB-CF	0.4991	0.6150	0.4282	0.5305	0.3977	0.4917	0.3788	0.4697
FB-MF	0.4890	0.6027	0.4269	0.5289	0.4019	0.4968	0.3864	0.4791
FB-EM	0.4856	0.5984	0.4195	0.5196	0.3925	0.4852	0.3751	0.4651