文献类型：期刊文献

中文题名：面向远程监督命名实体识别的噪声检测

英文题名：Noise Detection for Distant Supervised Named Entity Recognition

作者：王嘉诚[1];王凯[1];王昊奋[2];杜渂[3];何之栋[3];阮彤[1];刘井平[1]

机构：[1]华东理工大学信息科学与工程学院,上海200237;[2]同济大学设计与创意学院,上海200092;[3]迪爱斯信息技术股份有限公司,上海200032

年份：2024

卷号：61

期号：4

起止页码：916

中文期刊名：计算机研究与发展

外文期刊名：Journal of Computer Research and Development

收录：CSTPCD;;EI(收录号：20241816003593);Scopus;北大核心:【北大核心2023】；CSCD:【CSCD2023_2024】；

基金：上海市促进产业高质量发展专项资金(2021-GZL-RGZN-01018);国家重点研发计划项目(2021YFC2701800,2021YFC2701801);之江实验室开放课题(2019ND0AB01);上海市青年科技英才扬帆计划项目(23YF1409400)。

语种：中文

中文关键词：命名实体识别;远程监督;深度强化学习;噪声检测;预训练策略

外文关键词：named entity recognition;distant supervision;deep reinforcement learning;noise detection;pre-training strategy

摘要：针对远程监督命名实体识别(named entity recognition,NER)任务,目前有许多基于强化学习的方法,利用强化学习的强大决策能力,对远程监督生成的自动标注数据进行噪声过滤.然而,这些方法所使用的策略网络模型架构都较简单,识别噪声能力较弱,且都以完整的句子样本为单位进行识别,导致句子中的部分正确信息被丢弃.为解决上述问题,提出了一种新的基于强化学习的方法,称为RLTL-DSNER,该方法可以从远程监督生成的带噪数据中,以单词级别识别正确实例,减少噪声实例对远程监督NER的负面影响.具体来说,在策略网络模型中引入了标签置信函数来准确识别实例.此外,提出了一种新颖的NER模型预训练策略,使其能为强化学习的初始训练提供精准的状态表示和有效的奖励值,引导其向正确的方向更新.在4个数据集上的实验结果验证了RLTL-DSNER方法的优越性,在NEWS数据集上,相较于现有最先进的方法,获得了4.28%的F1提升.
On distantly supervised named entity recognition(NER),there are many reinforcement learning based approaches,which exploit the powerful decision-making ability of reinforcement learning to detect noise from the automatically labeled data generated by distant supervision.However,the structures of the policy network models used are typically simple,which results in a weak ability to recognize noisy instances.Furthermore,correct instances are identified at sentence level,resulting in part of the useful information in the sentence being discarded.In this paper,we propose a new reinforcement learning based method for distantly supervised NER,named RLTL-DSNER,which can detect correct instances at token level from noisy data generated by distant supervision,proposing to reduce the negative impact of noisy instances on distantly supervised NER model.Specifically,we introduce a tag confidence function to identify correct instances accurately.In addition,we propose a novel pretraining strategy for the NER model.This strategy can provide accurate state representations and effective reward values for the initial training of the reinforcement learning model.The pre-training strategy can help guide it to update in the right direction.We conduct experiments on four datasets to verify the superiority of the RLTL-DSNER method,gaining 4.28%F1 improvement on NEWS dataset over state-of-the-art methods.