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

Table 5

Accuracy comparison (a smaller value means higher accuracy).

Model	Training set density (TD) — throughput
	MAE	NMAE	MAE	NMAE	MAE	NMAE	MAE	NMAE
	UserMean	50.937	1.1729	51.343	1.1684	50.941	1.1676	51.185	1.1639
ItemMean	37.307	0.8597	33.014	0.7508	32.785	0.7490	32.676	0.7421
SlopeOne	31.798	0.7322	31.820	0.7242	31.688	0.7263	31.701	0.7208
IPCC	31.112	0.7164	29.936	0.6813	30.100	0.6899	30.609	0.6960
UPCC	30.829	0.7099	29.054	0.6612	28.357	0.6499	28.114	0.6393
WSRec	29.538	0.6802	28.185	0.6414	27.556	0.6315	27.422	0.6235
MF	58.623	1.3503	30.188	0.6870	24.106	0.5525	22.065	0.5017
FB-CF	24.189	0.5568	19.396	0.4413	17.522	0.4015	16.192	0.3682
FB-MF	21.870	0.5035	18.597	0.4231	17.533	0.4018	16.940	0.3852
FB-EM	21.806	0.5020	18.151	0.4130	16.884	0.3869	16.036	0.3646