摘要
[目的]为实现碳中和数据间的语义关联性挖掘、提升整体三元组抽取准确性,提出一种基于MacBERT的实体关系联合抽取HmBER模型.[方法]在HmBER模型中,通过相似度度量、实体边界辅助训练以及在关系抽取中引入实体类别特征,提升碳中和实体关系联合抽取的性能.[结果]与Multi-head、CasRel、SpERT和STER模型结果对比表明,HmBER模型的F1值在碳中和数据集上分别平均提升2.39%、13.84%.[局限]本方法处理的数据需要通过句子的意义推测实体关系联合抽取结果,没有做更深潜在语义的挖掘.[结论]HmBER模型有效地解决数据漏标与实体边界错误问题,为实体关系联合抽取提供了高准确抽取思路.
[Objective]To achieve semantic association mining between carbon-neutral data and improve the overall accuracy of triplet extraction,this paper proposes a HmBER model for joint entity-relation extraction based on MacBERT.[Methods]In the HmBER model,we enhanced the performance in joint extraction of carbon-neutral entity relationships through similarity measurement,auxiliary training with entity boundaries,and introducing entity category features in relation extraction.[Results]Compared with Multi-head,CasRel,SpERT,and STER models,the F1 score of the HmBER model on the carbon-neutral dataset achieved an average improvement by 2.39%and 13.84%,respectively.[Limitations]The data processed by this method requires inference of sentence meaning to derive entity-relation joint extraction results and deeper latent semantic mining was not performed.[Conclusions]The HmBER model effectively addresses data annotation omission and entity boundary errors,providing a highly accurate approach for entity-relation joint extraction.
作者
朱西平
肖丽娟
高昂
郭露
杨欢
Zhu Xiping;Xiao Lijuan;Gao Ang;Guo Lu;YangHuan(School of Electrical Engineering and Information,Southwest Petroleum University,Chengdu 610500,China)
出处
《数据分析与知识发现》
CSSCI
CSCD
北大核心
2024年第11期126-135,共10页
Data Analysis and Knowledge Discovery
基金
四川省科技计划基金项目(项目编号:2022YFQ0102)的研究成果之一。
关键词
实体关系联合抽取
碳中和
边界错误
数据漏标
Entity and Relation Extraction
Carbon Neutrality
Boundary Error
Missing Label
作者简介
通讯作者:肖丽娟,ORCID:0009-0002-8503-536X,E-mail:xiaojasminejuan@163.com。
