基于BiLSTM-CRF模型识别细密地址的特定标注方法分析
Analysis of Specific Annotation Methods for Identifying Fine-grained Addresses Based on BiLSTM-CRF model
邱坚 1张润熙 1沈俊杰1
作者信息
- 1. 中国移动上海分公司网络优化中心,上海 200060
- 折叠
摘要
阐述针对人为地址输入进行识别,通过深度学习将语义相似相近的地址完成实体名识别.以BiLSTM-CRF进行数据集的训练,针对前后文联系确定最佳可能出现地址,将海量地址进行特征提取和分类构造,得出多数F1指标91％以上.模型应用于实际数据中,验证了该方法的有效性.
Abstract
This paper describes the recognition of human input addresses,using deep learning to recognize entity names of semantically similar addresses.Train the dataset with BiLSTM-CRF,determine the best possible address based on the previous and subsequent connections,extract features and classify massive addresses,and obtain a majority of F1 indicators above 91％.The model was applied to actual data to verify the effectiveness of the method.
关键词
命名体识别/长短时记忆网络/条件随机场/信息抽取Key words
named entity recognition/long short-term memory network/conditional random field/information extraction引用本文复制引用
出版年
2024
NETL
NSTL
万方数据