/ChineseNER

Code for "Cross-Domain and Semi-Supervised Named Entity Recognition in Chinese Social Media: A Unified Model"

Primary LanguagePython

ChineseNER

Code for "Cross-Domain and Semi-Supervised Named Entity Recognition in Chinese Social Media: A Unified Model"

Requirements

  • ubuntu 16.04
  • python 2.7
  • theano 0.9.0

The processed data can be found at "data" folder. If you want to run our code on your own dataset, you first need to process your dataset into the specific data structure.

Run

bash run_train.sh
bash run_test.sh

Cite

To use this code, please cite the following paper:

Jingjing Xu, Hangfeng He, Xu Sun*, Xuancheng Ren, Sujian Li. Cross-Domain and Semi-Supervised Named Entity Recognition in Chinese Social Media: A Unified Model.  IEEE Transactions on Audio, Speech and Language Processing (TASLP) 2018.

bibtext:

@article{ChineseNER,
  author    = {Jingjing Xu and Hangfeng He and Xu Sun and Xuancheng Ren and Sujian Li},
  title     = {Cross-Domain and Semi-Supervised Named Entity Recognition in Chinese Social Media: A Unified Model},
  journal = {IEEE Transactions on Audio, Speech and Language Processing (TASLP)},
  year      = {2018}
}