Transfer Rule Learning over Large Knowledge Graphs

Hong Liu, Zhe Wang, Kewen Wang, Xiaowang Zhang, Zhiyong Feng

Track: Semantics and knowledge

Keywords: knowledge graph, transfer learning, rule learning

Abstract: Logical rules have been widely used for expressing schema knowledge in various practical applications. It is infeasible to handcraft rules from large knowledge graphs (KGs) and thus many methods have been proposed for learning rules automatically from KGs. However, it is largely ignored how to extract rules in a (target) KG from rules that already exist in some other (source) KGs. In this paper, we propose a framework for KG rule learning based on transfer learning. A major challenge for establishing such a framework is in that a suitable alignment mechanism is required for mapping certain subgraph structures between predicates in the source KG and the target KG. Hence, our framework provides a new method for predicate mapping based on the graph-structural similarity and thus, rules in the source KG can be transferred to the target KG. As not all transferred rules are valid ones in the target KG, methods are developed for further rule evaluation. The proposed framework can be used as a standalone rule learner but more importantly, it paves a new way for enhancing the state-of-the-art rule learners for large KGs. Extensive experiments are conducted to evaluate the new approach to rule learning, which show that rules in smaller KGs can be effectively transferred to a large KG.

Submission Number: 1342