Efficient complementary graph convolutional network without negative sampling for item recommendation

作者：

Highlights：

• Proposing an efficient complementary graph convolutional network for item recommendation.

• Designing an item-specific attention mechanism to fuse item representations from different bipartite graphs.

• Devising an efficient non-sampling learning paradigm to optimize the model parameters.

• Evaluating on six public benchmarks to demonstrate the effectiveness and efficiency of our solution.

摘要

•Proposing an efficient complementary graph convolutional network for item recommendation.•Designing an item-specific attention mechanism to fuse item representations from different bipartite graphs.•Devising an efficient non-sampling learning paradigm to optimize the model parameters.•Evaluating on six public benchmarks to demonstrate the effectiveness and efficiency of our solution.

论文关键词：Graph neural network,Attention mechanism,Complementary relationships,Recommender systems

论文评审过程：Received 14 December 2021, Revised 15 June 2022, Accepted 18 August 2022, Available online 27 August 2022, Version of Record 14 September 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2022.109758