Knowledge graph representation learning: A comprehensive and experimental overview

IF 13.3 1区 计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS Computer Science Review Pub Date : 2024-12-19 DOI:10.1016/j.cosrev.2024.100716
Dorsaf Sellami, Wissem Inoubli, Imed Riadh Farah, Sabeur Aridhi
{"title":"Knowledge graph representation learning: A comprehensive and experimental overview","authors":"Dorsaf Sellami, Wissem Inoubli, Imed Riadh Farah, Sabeur Aridhi","doi":"10.1016/j.cosrev.2024.100716","DOIUrl":null,"url":null,"abstract":"Knowledge graph embedding (KGE) is a hot topic in the field of Knowledge graphs (KG). It aims to transform KG entities and relations into vector representations, facilitating their manipulation in various application tasks and real-world scenarios. So far, numerous models have been developed in KGE to perform KG embedding. However, several challenges must be addressed when designing effective KGE models. The most discussed challenges in the literature include scalability (KGs contain millions of entities and relations), incompleteness (missing links), the complexity of relations (symmetries, inversion, composition, etc.), and the sparsity of some entities and relations. The purpose of this paper is to provide a comprehensive overview of KGE models. We begin with a theoretical analysis and comparison of the existing methods proposed so far for generating KGE, which we have classified into four categories. We then conducted experiments using four benchmark datasets to compare the efficacy, efficiency, inductiveness, the electricity and the CO<mml:math altimg=\"si1.svg\" display=\"inline\"><mml:msub><mml:mrow></mml:mrow><mml:mrow><mml:mn>2</mml:mn></mml:mrow></mml:msub></mml:math> emission of five state-of-the-art methods in the link prediction task, providing a comprehensive analysis of the most commonly used benchmarks in the literature.","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"81 1","pages":""},"PeriodicalIF":13.3000,"publicationDate":"2024-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computer Science Review","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.1016/j.cosrev.2024.100716","RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, INFORMATION SYSTEMS","Score":null,"Total":0}
引用次数: 0

Abstract

Knowledge graph embedding (KGE) is a hot topic in the field of Knowledge graphs (KG). It aims to transform KG entities and relations into vector representations, facilitating their manipulation in various application tasks and real-world scenarios. So far, numerous models have been developed in KGE to perform KG embedding. However, several challenges must be addressed when designing effective KGE models. The most discussed challenges in the literature include scalability (KGs contain millions of entities and relations), incompleteness (missing links), the complexity of relations (symmetries, inversion, composition, etc.), and the sparsity of some entities and relations. The purpose of this paper is to provide a comprehensive overview of KGE models. We begin with a theoretical analysis and comparison of the existing methods proposed so far for generating KGE, which we have classified into four categories. We then conducted experiments using four benchmark datasets to compare the efficacy, efficiency, inductiveness, the electricity and the CO2 emission of five state-of-the-art methods in the link prediction task, providing a comprehensive analysis of the most commonly used benchmarks in the literature.
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
知识图谱嵌入(KGE)是知识图谱(KG)领域的一个热门话题。它旨在将知识图谱中的实体和关系转化为矢量表示,以方便在各种应用任务和现实世界场景中对它们进行操作。迄今为止,KGE 已开发出许多模型来执行 KG 嵌入。然而,在设计有效的 KGE 模型时,必须解决几个难题。文献中讨论最多的挑战包括可扩展性(KG 包含数百万个实体和关系)、不完整性(缺失链接)、关系的复杂性(对称、反转、组合等)以及某些实体和关系的稀疏性。本文旨在全面概述 KGE 模型。我们首先对迄今为止提出的生成 KGE 的现有方法进行了理论分析和比较,并将其分为四类。然后,我们使用四个基准数据集进行了实验,比较了五种最先进方法在链接预测任务中的功效、效率、归纳性、电量和二氧化碳排放量,对文献中最常用的基准进行了全面分析。
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
Computer Science Review
Computer Science Review Computer Science-General Computer Science
CiteScore
32.70
自引率
0.00%
发文量
26
审稿时长
51 days
期刊介绍: Computer Science Review, a publication dedicated to research surveys and expository overviews of open problems in computer science, targets a broad audience within the field seeking comprehensive insights into the latest developments. The journal welcomes articles from various fields as long as their content impacts the advancement of computer science. In particular, articles that review the application of well-known Computer Science methods to other areas are in scope only if these articles advance the fundamental understanding of those methods.
期刊最新文献
A comprehensive survey of Federated Intrusion Detection Systems: Techniques, challenges and solutions Knowledge graph representation learning: A comprehensive and experimental overview A comprehensive review of usage control frameworks Integrating Explainable AI with Federated Learning for Next-Generation IoT: A comprehensive review and prospective insights Cloud continuum testbeds and next-generation ICTs: Trends, challenges, and perspectives
×
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