Multilingual Named Entity Recognition Using Pretrained Embeddings, Attention Mechanism and NCRF

Code

• github.com/sberbank-ai/ner-bert
  pytorch★ 0
• github.com/king-menin/slavic-ner
  pytorch★ 0

Abstract

In this paper we tackle multilingual named entity recognition task. We use the BERT Language Model as embeddings with bidirectional recurrent network, attention, and NCRF on the top. We apply multilingual BERT only as embedder without any fine-tuning. We test out model on the dataset of the BSNLP shared task, which consists of texts in Bulgarian, Czech, Polish and Russian languages.

Tasks

Reproductions