Analysis of influence of graph characteristics in the task of link prediction

О.П. Мосалов, А.О. Ефимова, Д.А. Соколова
{"title":"Analysis of influence of graph characteristics in the task of link prediction","authors":"О.П. Мосалов, А.О. Ефимова, Д.А. Соколова","doi":"10.21499/2409-1650-29-3-83-94","DOIUrl":null,"url":null,"abstract":"В работе рассматривается задача предсказания новых зависимостей между сущностями предметной области на основе информации, извлечённой из плохо структурированных данных. Предложены структура хранения таких данных в виде графа и алгоритм поиска новых сущностей на основе значений ряда характеристик такого графа. Приведены описание вычислительных экспериментов и обсуждение их результатов. Сформулированы требования к единой метрике оценки эффективности таких алгоритмов, приведены результаты анализа существующих метрик.\n In the article a task of link prediction in a subject area, that is based on the information extracted from improperly structured data, is considered. A structure to store such data in a form of a graph and an algorithm for link prediction based on the characteristics of the graph are suggested. The computational experiments are described and their results are discussed. Requirements for a unified metric for such algorithms are defined, and the results of existing metrics analysis are given.","PeriodicalId":424160,"journal":{"name":"Informacionno-technologicheskij vestnik","volume":"93 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Informacionno-technologicheskij vestnik","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.21499/2409-1650-29-3-83-94","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

В работе рассматривается задача предсказания новых зависимостей между сущностями предметной области на основе информации, извлечённой из плохо структурированных данных. Предложены структура хранения таких данных в виде графа и алгоритм поиска новых сущностей на основе значений ряда характеристик такого графа. Приведены описание вычислительных экспериментов и обсуждение их результатов. Сформулированы требования к единой метрике оценки эффективности таких алгоритмов, приведены результаты анализа существующих метрик. In the article a task of link prediction in a subject area, that is based on the information extracted from improperly structured data, is considered. A structure to store such data in a form of a graph and an algorithm for link prediction based on the characteristics of the graph are suggested. The computational experiments are described and their results are discussed. Requirements for a unified metric for such algorithms are defined, and the results of existing metrics analysis are given.
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
图特征在链路预测任务中的影响分析
本文探讨了根据从结构不良的数据中提取的信息预测学科领域实体之间新的依赖关系的问题。我们提出了一种以图表形式存储此类数据的结构,以及一种根据图表的若干特征值搜索新实体的算法。我们对计算实验进行了描述,并对实验结果进行了讨论。文章提出了对评估此类算法效率的单一指标的要求,并给出了对现有指标的分析结果。文章考虑了基于从结构不当的数据中提取的信息进行主题领域链接预测的任务。文章提出了一种以图表形式存储此类数据的结构,以及一种基于图表特征的链接预测算法。对计算实验进行了描述,并讨论了实验结果。定义了此类算法对统一度量的要求,并给出了现有度量分析的结果。
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
自引率
0.00%
发文量
0
期刊最新文献
Assessment of the Readiness of the Equipment of the Launch Space Complex at the Early Stages of its Creation The accuracy of measuring the information parameter of the signal by a tracking meter in aviation and rocket and space technology against the background of additive and multiplicative interference. Measurement of arrival time. Part III Problems of multiple use of liquid rocket propulsion systems Multiple sun-synchronous orbits for full coverage of the Earth Numerical and expert-oriented method of choosing the parameters of Sun-synchronous orbits for wide-scale monitoring Northern Russia regions
×
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