OD-Mind: An ocean drilling expert knowledge query system driven by knowledge graph

IF 4.6 2区工程技术 Q1 ENGINEERING, CIVIL

Ocean Engineering Pub Date : 2025-07-16 DOI:10.1016/j.oceaneng.2025.122135

Yulong Yang , Weihua Cao , Yupeng Li , Runzhou Chang , Gangcheng Yang , Chao Gan , Min Wu

{"title":"OD-Mind: An ocean drilling expert knowledge query system driven by knowledge graph","authors":"Yulong Yang , Weihua Cao , Yupeng Li , Runzhou Chang , Gangcheng Yang , Chao Gan , Min Wu","doi":"10.1016/j.oceaneng.2025.122135","DOIUrl":null,"url":null,"abstract":"<div><div>The textual publications generated by the Scientific Ocean Drilling compile a substantial body of knowledge. However, these publications present significant analytical challenges due to their technical language and rapidly expanding volume. To address these challenges, this study presents Ocean Drilling Mind (OD-Mind), a comprehensive system designed to automatically extract key knowledge from large-scale publications and support question answering in the Scientific Ocean Drilling domain. The contributions are threefold: 1) A knowledge extraction method is adapted to capture the knowledge related to complex multi-word compound named entities in ocean drilling; 2) A knowledge graph refinement method is proposed to consolidate the weak-connected knowledge into a high-density ocean drilling knowledge graph; 3) A query system is designed to answer specialized queries in the domain of ocean drilling based on the knowledge graph. In the case study, OD-Mind demonstrated its potential by constructing a comprehensive knowledge graph on 116 publications. Preliminary results showed that the system performed better than general-purpose large language models, particularly in answering specialized ocean drilling domain queries with improved speed and accuracy.</div></div>","PeriodicalId":19403,"journal":{"name":"Ocean Engineering","volume":"339 ","pages":"Article 122135"},"PeriodicalIF":4.6000,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ocean Engineering","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0029801825018190","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, CIVIL","Score":null,"Total":0}

引用次数: 0

Abstract

The textual publications generated by the Scientific Ocean Drilling compile a substantial body of knowledge. However, these publications present significant analytical challenges due to their technical language and rapidly expanding volume. To address these challenges, this study presents Ocean Drilling Mind (OD-Mind), a comprehensive system designed to automatically extract key knowledge from large-scale publications and support question answering in the Scientific Ocean Drilling domain. The contributions are threefold: 1) A knowledge extraction method is adapted to capture the knowledge related to complex multi-word compound named entities in ocean drilling; 2) A knowledge graph refinement method is proposed to consolidate the weak-connected knowledge into a high-density ocean drilling knowledge graph; 3) A query system is designed to answer specialized queries in the domain of ocean drilling based on the knowledge graph. In the case study, OD-Mind demonstrated its potential by constructing a comprehensive knowledge graph on 116 publications. Preliminary results showed that the system performed better than general-purpose large language models, particularly in answering specialized ocean drilling domain queries with improved speed and accuracy.

查看原文本刊更多论文

OD-Mind：一个基于知识图谱驱动的海洋钻井专家知识查询系统

科学海洋钻探产生的文本出版物汇编了大量的知识。然而，这些出版物由于其技术语言和迅速扩大的数量而提出了重大的分析挑战。为了应对这些挑战，本研究提出了Ocean Drilling Mind (OD-Mind)，这是一个综合系统，旨在自动从大型出版物中提取关键知识，并支持科学海洋钻探领域的问答。贡献体现在三个方面：1)提出了一种知识提取方法，用于捕获海洋钻井中复杂的多词复合命名实体相关知识；2)提出了一种知识图细化方法，将弱连接知识整合为高密度海洋钻井知识图；3)设计了一个基于知识图谱的查询系统，用于回答海洋钻井领域的专业查询。在案例研究中，OD-Mind通过构建116种出版物的综合知识图谱展示了其潜力。初步结果表明，该系统比通用大型语言模型表现更好，特别是在回答专门的海洋钻探领域查询时，速度和准确性都有所提高。

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分钟内获得全文求助全文

来源期刊

Ocean Engineering 工程技术-工程：大洋

CiteScore

7.30

自引率

34.00%

发文量

2379

审稿时长

8.1 months

期刊介绍： Ocean Engineering provides a medium for the publication of original research and development work in the field of ocean engineering. Ocean Engineering seeks papers in the following topics.