Research on knowledge graph alignment model based on deep learning.

• We propose a novel knowledge graph alignment model based on deep learning. • The experiments reveal that our model has effectively improved the performance. • We propose a novel negative sampling method, i.e., transformation negative sampling. • We examine the key influencing factors of knowledge graph alignment. • Our research has practical implications for improving the alignment performance. The construction of large-scale knowledge graphs from heterogeneous sources is fundamental to knowledge-driven applications. To solve the problem of redundancy and inconsistency in the process of domain knowledge fusion, this paper reports studies of domain knowledge alignment from the perspective of a knowledge graph. A novel knowledge graph alignment (KGA) model is proposed, based on knowledge graph deep representation learning. To assess the validity of the model, comparative experiments are conducted on the datasets of heterogeneous, cross-lingual, and domain-specific knowledge graphs. Our results of experiments suggest significant improvement on all of these datasets. We discuss the implications for improving the alignment effect of knowledge graph entities, enhancing the coverage and correctness of knowledge graphs, and promoting the performance of knowledge graphs in knowledge-driven applications. [ABSTRACT FROM AUTHOR]