Chinese NER based on improved Transformer encoder

Abstract

In order to improve the effect of chinese named entity recognition, a method based on the XLNET- Transformer_P-CRF model was proposed, which used the Transformer_P encoder, improved the shortcomings of the traditional Transformer encoder that couldn’t obtain relative position information. Experiments show that the XLNET-Transformer_P-CRF model achieves 95.11%, 80.54%, 96.70%, and 71.46% F1 values on the four types of data sets: MSRA, OntoNotes4.0, Resume, and Weibo, which are all higher than other mainstream chinese NER model.

Keywords

• chinese named entity recognition
• transformer encoder
• relative position information