On the Form of Parsed Sentences for Relation Extraction

Xiaoying Chen,Mi Zhang,Shengwu Xiong,Tieyun Qian

DOI: https://doi.org/10.1016/j.knosys.2022.109184

IF: 8.139

2022-01-01

Knowledge-Based Systems

Abstract:Parsed sentences convey rich syntactic information like part-of-speech (POS) tags and dependency trees, and have been widely adopted in relation extraction. However, existing methods either suffer from error propagation when using tree or graph form of the imperfect parse tree, or neglect the independent POS sequence because each POS embedding is combined with a word embedding to form the representation of the word. We propose to exploit the sequential form of POS tags beyond the popular tree or graph form of parse tree of a sentence. Our method naturally fills the gap between the original sentence and imperfect parse tree. It also enables the learnt POS embeddings to correspond and interact with word embeddings pre-trained by sequential models like GloVe or BERT. This property is further leveraged in a downstream entity masking task designed for relation extraction. We conduct extensive experiments on various type of relation extraction tasks. The results demonstrate that our model significantly outperforms the state-of-the-art approaches.