NeuRec: On Nonlinear Transformation for Personalized Ranking

05/08/2018

∙

by   Shuai Zhang, et al.

∙

0

∙

share

Modelling user-item interaction patterns is an important task for personalized recommendations. Many recommender systems are based on the assumption that there exists a linear relationship between users and items, while neglecting the intricacy and non-linearity of real-life historical interactions. In this paper, we propose a neural recommendation (NeuRec) model that untangles the complexity of user-item interactions and establish an integrated network to link a non-linear neural transformation part and latent factor part. To explore its capability, we design two variants of NeuRec: user based NeuRec (U-NeuRec) and item based NeuRec (I-NeuRec). Extensive experiments on four real-world datasets demonstrated its superior performances on personalized ranking task.