Relation Extraction Based on Prompt Information and Feature Reuse

IF 1.3 3区 计算机科学 Q3 COMPUTER SCIENCE, INFORMATION SYSTEMS Data Intelligence Pub Date : 2023-02-22 DOI:10.1162/dint_a_00192
Ping Feng, Xin Zhang, Jian Zhao, Yingying Wang, Biao Huang
{"title":"Relation Extraction Based on Prompt Information and Feature Reuse","authors":"Ping Feng, Xin Zhang, Jian Zhao, Yingying Wang, Biao Huang","doi":"10.1162/dint_a_00192","DOIUrl":null,"url":null,"abstract":"ABSTRACT To alleviate the problem of under-utilization features of sentence-level relation extraction, which leads to insufficient performance of the pre-trained language model and underutilization of the feature vector, a sentence-level relation extraction method based on adding prompt information and feature reuse is proposed. At first, in addition to the pair of nominals and sentence information, a piece of prompt information is added, and the overall feature information consists of sentence information, entity pair information, and prompt information, and then the features are encoded by the pre-trained language model ROBERTA. Moreover, in the pre-trained language model, BIGRU is also introduced in the composition of the neural network to extract information, and the feature information is passed through the neural network to form several sets of feature vectors. After that, these feature vectors are reused in different combinations to form multiple outputs, and the outputs are aggregated using ensemble-learning soft voting to perform relation classification. In addition to this, the sum of cross-entropy, KL divergence, and negative log-likelihood loss is used as the final loss function in this paper. In the comparison experiments, the model based on adding prompt information and feature reuse achieved higher results of the SemEval-2010 task 8 relational dataset.","PeriodicalId":34023,"journal":{"name":"Data Intelligence","volume":"5 1","pages":"824-840"},"PeriodicalIF":1.3000,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Data Intelligence","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.1162/dint_a_00192","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, INFORMATION SYSTEMS","Score":null,"Total":0}
引用次数: 1

Abstract

ABSTRACT To alleviate the problem of under-utilization features of sentence-level relation extraction, which leads to insufficient performance of the pre-trained language model and underutilization of the feature vector, a sentence-level relation extraction method based on adding prompt information and feature reuse is proposed. At first, in addition to the pair of nominals and sentence information, a piece of prompt information is added, and the overall feature information consists of sentence information, entity pair information, and prompt information, and then the features are encoded by the pre-trained language model ROBERTA. Moreover, in the pre-trained language model, BIGRU is also introduced in the composition of the neural network to extract information, and the feature information is passed through the neural network to form several sets of feature vectors. After that, these feature vectors are reused in different combinations to form multiple outputs, and the outputs are aggregated using ensemble-learning soft voting to perform relation classification. In addition to this, the sum of cross-entropy, KL divergence, and negative log-likelihood loss is used as the final loss function in this paper. In the comparison experiments, the model based on adding prompt information and feature reuse achieved higher results of the SemEval-2010 task 8 relational dataset.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
基于提示信息和特征重用的关系提取
摘要为了解决句子级关系提取中特征利用不足,导致预先训练的语言模型性能不足和特征向量利用不足的问题,提出了一种基于添加提示信息和特征重用的句子级关系抽取方法。首先,除了名词对和句子信息外,还添加了一条提示信息,整体特征信息由句子信息、实体对信息和提示信息组成,然后通过预先训练的语言模型ROBERTA对特征进行编码。此外,在预训练的语言模型中,在神经网络的组成中也引入了BIGRU来提取信息,并将特征信息通过神经网络形成多组特征向量。之后,将这些特征向量以不同的组合重复使用,以形成多个输出,并使用集成学习软投票对输出进行聚合,以执行关系分类。除此之外,本文还使用交叉熵、KL散度和负对数似然损失之和作为最终损失函数。在对比实验中,基于添加提示信息和特征重用的模型在SemEval-2010任务8关系数据集中获得了更高的结果。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Data Intelligence
Data Intelligence COMPUTER SCIENCE, INFORMATION SYSTEMS-
CiteScore
6.50
自引率
15.40%
发文量
40
审稿时长
8 weeks
期刊最新文献
The Limitations and Ethical Considerations of ChatGPT Rule Mining Trends from 1987 to 2022: A Bibliometric Analysis and Visualization Classification and quantification of timestamp data quality issues and its impact on data quality outcome BIKAS: Bio-Inspired Knowledge Acquisition and Simulacrum—A Knowledge Database to Support Multifunctional Design Concept Generation Exploring Attentive Siamese LSTM for Low-Resource Text Plagiarism Detection
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1