Hierarchical Relation Extraction Based On Bidirectional Long Short-Term Memory Networks

Abstract

Relation extraction is an important task in the field of natural language processing (NLP). Most of the present methods extract each relation in isolation, without considering the hierarchical semantic information between relations. A novel loss function to optimize model of relation extraction based on hierarchical relation has been proposed in this paper. The experimental results show that the proposed model outperforms most of the present methods.