From: Click-through rate prediction model integrating user interest and multi-head attention mechanism
Model | Criteo | |||||
---|---|---|---|---|---|---|
Proportion | 0.2 | 0.3 | AVG. changes | |||
 | AUC | LOSS | AUC | LOSS | AUC | LOSS |
FM | 0.6869 | 0.5286 | 0.6785 | 0.5368 | − 0.0084 |  + 0.0082 |
Weidedeep | 0.7066 | 0.4827 | 0.7007 | 0.4846 | − 0.0059 |  + 0.0019 |
Deepfm | 0.7283 | 0.4707 | 0.7249 | 0.4792 | − 0.0034 |  + 0.0085 |
AFM | 0.7220 | 0.4754 | 0.7084 | 0.4877 | − 0.0136 |  + 0.0123 |
DCN | 0.7094 | 0.4920 | 0.7042 | 0.5010 | − 0.0052 |  + 0.009 |
NFM | 0.7027 | 0.5645 | 0.7005 | 0.5654 | − 0.0022 |  + 0.0009 |
PNN | 0.7084 | 0.4870 | 0.7013 | 0.4979 | − 0.0071 |  + 0.0109 |
Autoint | 0.7060 | 0.6049 | 0.6921 | 0.6552 | − 0.0139 |  + 0.0503 |
Deepcrosing | 0.7375 | 0.4732 | 0.7356 | 0.4792 | −0.0019 |  + 0.006 |
IARM | 0.7545 | 0.4830 | 0.7527 | 0.4993 | − 0.0018 |  + 0.0163 |