# daguan-information-extraction
“达观杯”文本智能信息抽取挑战赛
## 解决方案
- 001_crf_baseline
对原始数据进行标注，格式参照crf++提供的例子，然后使用crf++工具进行训练
- 003_daguan_named_entity_recognition
https://github.com/wenyu332/daguan_named_entity_recognition
- 004_xunlian_keras
未实现
https://github.com/HulkZhang001/ChineseNER/tree/master/xunlian_keras
- Named-Entity-Recognition-BidirectionalLSTM-CNN-CoNLL
https://github.com/mxhofer/Named-Entity-Recognition-BidirectionalLSTM-CNN-CoNLL
- jiesutd/LatticeLSTM
https://github.com/jiesutd/LatticeLSTM
“达观杯”文本智能信息抽取挑战赛.zip （48个子文件） 
资料总结 001_crf_baseline 
__init__.py 0B CRF_baseline.py 4KB template 224B submit.py 336B 000.make_ner_format_data.py 4KB 001.make_submit.py 3KB
readme.txt 134B 002_bilstm_crf predict_submit.py 7KB train_w2v.py 2KB train_bilstm_crf.py 2KB prepare_data.py 6KB __pycache__ prepare_data.cpython-37.pyc 6KB 004_xunlian_keras test.ipynb 8KB BiLSTM+CRF(renminRB).ipynb 11KB bilsm_crf_model.py 920B test_data.data 1.06MB val.py 895B config.pkl 53KB train_data.data 13.26MB BiLSTM+CRF(MSRA).ipynb 15KB BiLSTM+CRF(Boson).ipynb 12KB process_data.py 2KB train.py 244B __pycache__ bilsm_crf_model.cpython-37.pyc 1KB process_data.cpython-37.pyc 3KB README.md 624B 003_daguan_named_entity_recognition utils.py 7KB tag2id 162B corpus_word2id 360KB evaluate.py 1KB main.py 3KB word2id 77KB normal_daguan_test.txt 839KB models util.py 5KB bilstm_crf.py 14KB bilstm.py 3KB __pycache__ bilstm_crf.cpython-37.pyc 8KB config.cpython-37.pyc 623B bilstm.cpython-37.pyc 2KB util.cpython-37.pyc 3KB config.py 259B next_dev_result 1.453_bilstmcrf_result.txt 341KB result.txt 875KB normal_daguan_train.txt 6.48MB __pycache__ evaluate.cpython-37.pyc 1KB utils.cpython-37.pyc 5KB .gitignore 250B README.md 649B资源评论
