基于多任务标签一致性机制的中文命名实体识别  

Chinese Named Entity Recognition Based on Multi-task Label Consistency

作　　者：吕书宁 刘健 徐金安[1] 陈钰枫[1] 张玉洁[1] LYU Shuning;LIU Jian;XU Jin an;CHEN Yufeng;ZHANG Yujie(School of Computer and Information Technology,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]北京交通大学计算机与信息技术学院,北京100044

出　　处：《中文信息学报》2023年第12期87-97,共11页Journal of Chinese Information Processing

基　　金：国家重点研究与发展计划项目(2019YFB1405200);国家自然科学基金(61976015,61976016,61876198,61370130)。

摘　　要：实体边界预测对中文命名实体识别至关重要。现有研究为改善边界识别效果而提出的多任务学习方法大多仅考虑与分词任务进行简单结合,但由于缺少包含多任务标签的训练数据,导致无法学到多个任务之间的标签一致性关系。该文提出一种新的基于多任务标签一致性机制的中文命名实体识别方法:将分词和词性信息融入命名实体识别模型,进而联合训练命名实体识别、分词、词性标注三种任务;建立基于标签一致性机制的多任务学习模式,增强边界信息学习,捕获标签一致性关系,更好地学习多任务表示。相较于基线模型,全样本实验、模拟小样本实验及真实小样本实验分别提升F_1值10.28%、11.17%和8.84%,表明了该方法的有效性。Entity boundary prediction is essential for Chinese named entity recognition.Most of multi-task learning methods consider only employing the word segmentation task.This paper presents a new Chinese named entity recognition method based on the multi-task label consistency mechanism.The method integrates word segmentation and part-of-speech information into the named entity recognition model to establish a multi-task learning mode based on the label consistency mechanism,It enhances the boundary information learning by capturing label consistency relationships under the multi-task framework.The method is vilidated by the full sample experiment,simulated small sample experiment and real small sample experiment,resulting 10.28%,11.17%and 8.84%improvements over the baseline model,respectively.

关 键 词：中文命名实体识别 多任务学习 标签一致性机制 BERT模型 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]