Cross-Knowledge Graph Entity Alignment via Neural Tensor Network

Abstract With the expansion of current knowledge graph scale and increase number entities, a large graphs express same entity in different ways, so importance fusion is increasingly manifested. Traditional alignment algorithms have limited application scope low efficiency. This paper proposes an method based on neural tensor network (NtnEA), which can obtain inherent semantic information text without being restricted by linguistic features structural information, relying string information. In three cross-lingual language data sets DBP FR−EN , ZH−EN JP−EN DBP15K set, Mean Reciprocal Rank Hits@k are used as effect evaluation indicators for tasks. Compared with existing methods MTransE, IPTransE, AlignE AVR-GCN, Hit@10 values NtnEA 85.67, 79.20, 78.93, MRR 0.558, 0.511, 0.499, better than traditional improved 10.7% average.