Recommendation model based on separated embedding interaction networks

Aiming at the problems that the existing feature interaction methods in deep learning recommendation models cannot fully utilize the embedding vector information and thus have the insufficient accuracy, we propose a deep learning recommendation model based on separated embedding interaction networks (SEIN). This model first uses the embedding neural network layer to convert the sparse feature vectors into dense embedding vectors, then separates the feature matrices of different dimensions for feature interaction, and explicitly controls the order of feature interaction through the number of SEIN layers. Finally, the obtained hidden layer matrices are pooled by summation, and the final output is obtained through the prediction layer. In public datasets of Criteo, AutoML and Movielens, click-through rate prediction and top-k recommendation experiments are carried out by using the area under the curve (AUC), log-loss, accuracy and recall rate as evaluation indicators. The experimental results demonstrate that compared with the baseline models for click-through rate prediction, namely DeepFM, Deep&Cross Received: 202211-25; Accepted: 2023-04-02; Online (CNKI): 2023-06-05 Foundation: Construction Project of Applied Undergraduate Pilot Program in Shanghai Universities (Z32004-17-84); Special Project of Shanghai Municipal Education Commission for the Construction of First-Class Undergraduate Majors (JYLB202002) Corresponding author: Associate professor JIANG Zhongyun. E-mail: jianqiao_jzy@163.com Citation: FENG Shulei, JIANG Zhongyun. Recommendation model based on separated embedding interaction networks [J]. Journal of Shenzhen University Science and Engineering, 2023, 40(4): 513-520. (in Chinese)