A Hybrid Recommender System for Gaussian Mixture Model and Enhanced Social Matrix Factorization Technology Based on Multiple Interests

Table 3

Parameters combination performance on Epinions and Tencent datasets.

Epinions dataset	MAE	Tencent dataset	MAE
Parameters combination		Parameters combination
= 0.5, ω = 0.01, φ = 0	0.781	γ = 0.5, ω = 0.01, φ = 0	0.721
γ = 0.5, ω = 0.05, φ = 0.4	0.716	γ = 0.5, ω = 0.08, φ = 0.4	0.658
γ = 0.5, ω = 0.07, φ = 0.5	0.726	γ = 0.5, ω=0.09, φ = 0.5	0.672
γ = 2, ω = 0.01, φ = 0	0.697	γ = 2, ω = 0.01, φ = 0	0.691
γ = 2, ω = 0.05, φ = 0.4	0.665	γ = 2, ω = 0.08, φ = 0.4	0.628
γ = 2, ω = 0.05, φ = 0.5	0.671	γ = 2, ω = 0.09, φ = 0.5	0.641
γ = 2.5, ω = 0.01, φ = 0	0.722	γ = 2.5, ω = 0.01, φ = 0	0.679
γ = 2.5, ω = 0.05, φ = 0.4	0.656	γ = 2.5, ω = 0.08, φ = 0.4	0.602
γ = 2.5, ω = 0.07, φ = 0.5	0.668	γ = 2.5, ω = 0.09, φ = 0.5	0.627
γ = 3, ω = 0.01, φ = 0	0.721	γ = 3.5, ω = 0.01, φ = 0	0.681
γ = 3, ω = 0.05, φ = 0.4	0.662	γ = 3.5, ω = 0.08, φ = 0.4	0.616
γ = 3, ω = 0.07, φ = 0.5	0.672	γ = 3.5, ω = 0.09, φ = 0.5	0.629
γ = 3.5, ω = 0.01, φ = 0	0.699	γ = 4, ω = 0.01, φ = 0	0.691
γ = 3.5, ω = 0.05, φ = 0.4	0.670	γ = 4, ω = 0.08, φ = 0.4	0.618
γ = 3.5, ω = 0.07, φ = 0.5	0.675	γ = 4, ω = 0.09, φ = 0.5	0.633