GCKG: Novel Gated Convolutional embedding model for Knowledge Graphs

作者：

Highlights：

• Gated Mechanism for KGs is proposed considering characteristic of a triple.

• Gated Mechanism is inspired by the chain rule and the Markov assumption.

• A Gated convolution Embedding for KGs.

• The model achieved a good performance on benchmark datasets.

摘要

•Gated Mechanism for KGs is proposed considering characteristic of a triple.•Gated Mechanism is inspired by the chain rule and the Markov assumption.•A Gated convolution Embedding for KGs.•The model achieved a good performance on benchmark datasets.

论文关键词：Knowledge graphs,Knowledge embedding,Gated mechanism,KGC

论文评审过程：Received 20 April 2021, Revised 13 January 2022, Accepted 10 July 2022, Available online 19 July 2022, Version of Record 25 July 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118142