DeepFM: A Factorization-Machine based Neural Network for CTR Prediction

13 Mar 2017  ·  Huifeng Guo, Ruiming Tang, Yunming Ye, Zhenguo Li, Xiuqiang He ·

Learning sophisticated feature interactions behind user behaviors is critical in maximizing CTR for recommender systems. Despite great progress, existing methods seem to have a strong bias towards low- or high-order interactions, or require expertise feature engineering. In this paper, we show that it is possible to derive an end-to-end learning model that emphasizes both low- and high-order feature interactions. The proposed model, DeepFM, combines the power of factorization machines for recommendation and deep learning for feature learning in a new neural network architecture. Compared to the latest Wide \& Deep model from Google, DeepFM has a shared input to its "wide" and "deep" parts, with no need of feature engineering besides raw features. Comprehensive experiments are conducted to demonstrate the effectiveness and efficiency of DeepFM over the existing models for CTR prediction, on both benchmark data and commercial data.

PDF Abstract


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Click-Through Rate Prediction Amazon DeepFM AUC 0.8683 # 3
Click-Through Rate Prediction Bing News DeepFM AUC 0.8376 # 3
Log Loss 0.2671 # 3
Click-Through Rate Prediction Company* DeepFM AUC 0.8715 # 1
Log Loss 0.02618 # 1
Click-Through Rate Prediction Criteo DeepFM AUC 0.8007 # 31
Log Loss 0.45083 # 17
Click-Through Rate Prediction Dianping DeepFM AUC 0.8481 # 2
Log Loss 0.3333 # 2
Click-Through Rate Prediction MovieLens 20M DeepFM AUC 0.7324 # 4


No methods listed for this paper. Add relevant methods here