Publication - Paper

The adaptive margin in hTransA

Date

    2015/11 - 2016/02

Authors

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

Paper

Problem

    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, 50\% triples in knowledge graphs are organized in hierarchical structures, which also contain rich inference patterns to predict links but do not be fully used.

My Work

  • Proposed a hierarchy-constrained link prediction method by defining single-step specific margin, called hTransA, on the basis of the translation-based knowledge graph embedding methods. It can adaptively determine the optimal margin to separate positive and negative triples.
  • Provided one way to detect the hierarchical structures in knowledge graphs by employing the properties of hierarchical relations.
  • Experimented hTransA on two datasets, and it decreased the mean rank of the state-of-the-art method at that time, TransA, by more than 20%.
  • The paper was published as a poster by WWW 2016 (Lead Author).