Block term decomposition with distinct time granularities for temporal knowledge graph completion

作者：

Highlights：

• We encode the fine-grained and coarse-grained time information of facts through BTD.

• We explicitly model the static properties of entities and relations.

• Our model unifies the processing in handling discrete and continuous facts.

• Our model outperforms state-of-the-art methods on three datasets of different types.

摘要

•We encode the fine-grained and coarse-grained time information of facts through BTD.•We explicitly model the static properties of entities and relations.•Our model unifies the processing in handling discrete and continuous facts.•Our model outperforms state-of-the-art methods on three datasets of different types.

论文关键词：Knowledge graph completion,Temporal knowledge graph,Tensor completion

论文评审过程：Received 24 October 2021, Revised 18 January 2022, Accepted 27 March 2022, Available online 9 April 2022, Version of Record 19 April 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117036