Hyperparameter ranges for all considered baseline models

Algorithms topK shrink Similarity Normalize α β ℓ2 norm
kNN algorithms [5 to 1000] [0 to 1000] [cosine, jaccard, asymmetric, dice, tversky] [True, False] — — —
Graph based algorithms [5 to 1000] — — — [0 to 2], strategy: uniform distribution [0 to 2], strategy: uniform distribution —
EASER — — — — — — [1 to 1e7], strategy: log-uniform distribution


Optimal hyperparameter values

Algorithms topK shrink Similarity Normalize α β ℓ2 norm
ItemkNN 144 1000 cosine True — — —
UserkNN 146 1000 cosine True — — —
P3α 496 — — False 0.7681732734954694 — —
RP3β 496 — — True 0.44477903655656115 0.5968193614337285 —
EASER — — — — — — 1e3


Statistics of the provided train-test splits

Users Items Interactions Density Data leakage Interactions per usermin Interactions per itemmin Interactions per usermax Interactions per itemmax Testing users
52,643 91,599 2,984,108 0.00062 No 16 10 13313 1907 52,639


Full result table

Methods Amazon-Book Time(s)
R@20 N@20 T-Time P-Time Average P-Time Per User
Random 0.00022 0.00018 0 150.7472472 0.002863576
TopPop 0.0024 0.0044 0.046885014 128.3760633 0.002438616
ItemkNN 0.056 0.045 426.9121618 107.231256 0.002036952
UserkNN 0.051 0.041 283.7361674 104.065511 0.001976816
P3α 0.056 0.043 197.1278133 114.8666947 0.002181994
RP3β 0.068 0.056 224.2510846 126.1138134 0.002395643
EASER 0.071 0.056 — — —
DGCF (Paper) 0.039 0.030
DGCF (Code) 0.011 0.017 — — —

Optimal hyperparameter values

Algorithms topK shrink Similarity Normalize α β ℓ2 norm
ItemkNN 144 1000 cosine True — — —
UserkNN 146 1000 cosine True — — —
P3α 496 — — False 0.7681732734954694 — —
RP3β 350 — — True 0.7681732734954694 0.4181395996963926 —
EASER — — — — — — 1e3


Statistics of the provided train-test splits

Users Items Interactions Density Data leakage Interactions per usermin Interactions per itemmin Interactions per usermax Interactions per itemmax Testing users
31,668 38,048 1,561,406 0.00130 No 18 10 2311 1492 31,668


Full result table

Methods Yelp'18 Time(s)
R@20 N@20 T-Time P-Time Average P-Time Per User
Random 0.0003 0.0004 0 53.14148092 0.001678081
TopPop 0.0125 0.0101 0.006148577 45.95453548 0.001451135
ItemkNN 0.064 0.054 81.77051783 47.61694789 0.00150363
UserkNN 0.061 0.050 72.28791833 43.42841244 0.001371366
P3α 0.064 0.052 64.42336011 46.88240695 0.001480435
RP3β 0.067 0.056 54.6901195 48.22654772 0.001522879
EASER 0.064 0.052 555.2657838 84.11382508 0.002656114
DGCF (Paper) 0.064 0.052
DGCF (Code) 0.037 0.028 — — —

Optimal hyperparameter values

Algorithms topK shrink Similarity Normalize α β ℓ2 norm
ItemkNN 508 1000 cosine True — — —
UserkNN 146 1000 cosine True — — —
P3α 496 — — False 1.087096950563704 — —
RP3β 350 — — True 0.5663562161452378 0.001085447926739258 —
EASER — — — — — — 1e3


Statistics of the provided train-test splits

Users Items Interactions Density Data leakage Interactions per usermin Interactions per itemmin Interactions per usermax Interactions per itemmax Testing users
29,858 40,981 1,027,370 0.00084 No 9 10 1,014 2,310 29,858


Full result table

Methods Yelp'18 Time(s)
R@20 N@20 T-Time P-Time Average P-Time Per User
Random 0.00013 0.00022 0 51.75596523 0.001733404
TopPop 0.0139 0.0317 0 45.00996447 0.001507467
ItemkNN 0.156 0.123 89.02916408 49.55570626 0.001659713
UserkNN 0.155 0.128 58.89179349 42.49520731 0.001423244
P3α 0.175 0.146 90.30393124 48.97286344 0.001640192
RP3β 0.183 0.152 91.19598675 49.15153646 0.001646176
EASER 0.163 0.136 724.8426113 76.08247638 0.002548144
DGCF (Paper) 0.179 0.152
DGCF (Code) 0.133 0.109 — — —