Hierarchy-Based Link Prediction in Knowledge Graphs

Manling Li, Yantao Jia, Yuanzhuo Wang, Jingyuan Li, Xueqi Cheng

Abstract: Link prediction over a knowledge graph aims to predict the missing entity h or t for a triple (h,r,t). Existing knowledge graph embedding based predictive methods represent entities and relations in knowledge graphs as elements of a vector space, and employ the structural information for link prediction. However, knowledge graphs contain many hierarchical relations, which existing methods have pay little attention to. In this paper, we propose a hierarchy-constrained locally adaptive knowledge graph embedding based link prediction method, called hTransA, by integrating hierarchical structures into the predictive work. Experiments over two benchmark data sets demonstrate the superiority of hTransA.