Dessí Danilo, Osborne Francesco, Buscaldi Davide, Reforgiato Recupero Diego, Motta Enrico
Department of Computer Science, College of Computing and Informatics, University of Sharjah, Sharjah, UAE.
The Open University, Knowledge Media Institute, Milton Keynes, UK.
Sci Data. 2025 Jun 9;12(1):964. doi: 10.1038/s41597-025-05200-8.
The rapid evolution of AI and the increased accessibility of scientific articles through open access marks a pivotal moment in research. AI-driven tools are reshaping how scientists explore, interpret, and contribute to the body of scientific knowledge, offering unprecedented opportunities. Nonetheless, a significant challenge remains: dealing with the overwhelming number of papers published every year. A promising solution is the use of knowledge graphs, which provide structured, interconnected, and formalized frameworks that improve the capabilities of AI systems to integrate information from the literature. This paper presents the last version of the Computer Science Knowledge Graph (CS-KG 2.0), an extensive knowledge base generated from 15 million research papers. CS-KG 2.0 describes 25 million entities linked by 67 million relationships, offering a nuanced representation of the scientific knowledge within the field of computer science. This innovative resource facilitates new research opportunities in key areas such as analysis and forecasting of research trends, hypothesis generation, smart literature search, automatic production of literature review, and scientific question-answering.
人工智能的快速发展以及通过开放获取使科学文章更易获取，标志着研究领域的一个关键时刻。人工智能驱动的工具正在重塑科学家探索、解释和为科学知识体系做出贡献的方式，带来了前所未有的机遇。尽管如此，一个重大挑战仍然存在：应对每年发表的大量论文。一个有前景的解决方案是使用知识图谱，它提供结构化、相互关联且形式化的框架，可提高人工智能系统整合文献信息的能力。本文介绍了计算机科学知识图谱（CS-KG 2.0）的最新版本，这是一个由1500万篇研究论文生成的广泛知识库。CS-KG 2.0描述了由6700万个关系链接的2500万个实体，对计算机科学领域内的科学知识进行了细致入微的呈现。这种创新资源为研究趋势分析与预测、假设生成、智能文献搜索、文献综述自动生成以及科学问答等关键领域带来了新的研究机会。
