xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems

14 Mar 2018Jianxun LianXiaohuan ZhouFuzheng ZhangZhongxia ChenXing XieGuangzhong Sun

Combinatorial features are essential for the success of many commercial models. Manually crafting these features usually comes with high cost due to the variety, volume and velocity of raw data in web-scale systems... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Click-Through Rate Prediction Bing News xDeepFM AUC 0.8400 # 1
Click-Through Rate Prediction Bing News xDeepFM Log Loss 0.2649 # 1
Click-Through Rate Prediction Criteo xDeepFM AUC 0.8052 # 1
Click-Through Rate Prediction Criteo xDeepFM Log Loss 0.4418 # 1
Click-Through Rate Prediction Dianping xDeepFM AUC 0.8639 # 1
Click-Through Rate Prediction Dianping xDeepFM Log Loss 0.3156 # 1