Embedding ranking-oriented recommender system graphs

作者：

Highlights：

• PGRec is a novel graph-based ranking-oriented recommendation framework.

• PGRec models the user preferences, by a novel heterogeneous graph called PrefGraph.

• PGRec embeds nodes in graph with a hybrid factorization and deep learning method.

• Unknown preferences are predicted by applying Neural Network on node embeddings.

• PGRec outperforms the baseline methods in terms of the NDCG in several datasets.

摘要

•PGRec is a novel graph-based ranking-oriented recommendation framework.•PGRec models the user preferences, by a novel heterogeneous graph called PrefGraph.•PGRec embeds nodes in graph with a hybrid factorization and deep learning method.•Unknown preferences are predicted by applying Neural Network on node embeddings.•PGRec outperforms the baseline methods in terms of the NDCG in several datasets.

论文关键词：Ranking-oriented recommender system,Deep learning,Graph embedding,Convolution

论文评审过程：Received 14 June 2020, Revised 24 March 2021, Accepted 21 April 2021, Available online 23 April 2021, Version of Record 15 May 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115108