Skip to content

Files

Latest commit

24d532d · Jun 24, 2020

History

History
18 lines (14 loc) · 921 Bytes

README.md

File metadata and controls

18 lines (14 loc) · 921 Bytes

Chinese_Resume_NER

Using HMM,CRF,LSTM,LSTM+CRF to achieve the goal of name entity recognition.

Environment:
python3.6
tensorflow 1.12+

Refer to this github

Dataset is published by ACL 2018 Chinese NER using Lattice LSTM

This project don't contain the model saver, and the evaluation is about precision, recall and F1 of token pair, not about word, so you can complete these parts. The result of this project:

HMM CRF BiLSTM BiLSTM+CRF RoBERTa_small BERT-base RoBERTa-base
recall 90.93% 94.78% 93.96% 95.39% 96.34% 98.57% 98.58%
precision 93.13% 96.51% 96.41% 96.61% 100% 100% 100%
F1 score 91.97% 95.63% 95.13% 95.91% --- --- ---