PubMed knowledge graph 2.0: Connecting papers, patents, and clinical trials in biomedical science

2024-10-10

arXiv.org

AI Summary

Biomedical literature—including scholarly articles, patents, and clinical trials—is highly fragmented and heterogeneous, impeding fine-grained cross-source integration. To address this, we construct a unified knowledge graph covering 36 million articles, 1.3 million patents, and 480,000 clinical trials—the first to systematically link these three literature types at both the entity level and the NIH-funded project level, augmented with NIH project metadata to enhance knowledge provenance. Methodologically, we integrate fine-grained biomedical entity recognition, high-precision author name disambiguation, multi-source citation fusion, and scalable knowledge graph construction. Our approach achieves state-of-the-art performance on both author disambiguation and biomedical entity recognition benchmarks. The resulting open knowledge infrastructure significantly advances literature mining, research evaluation, and translational medicine support by enabling robust, cross-modal knowledge discovery and evidence tracing.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Knowledge Representation and Reasoning: Knowledge Engineering
• Machine Learning: Graph-based Machine Learning

Application Category

• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Web data integration and cleaning

Abstract

Papers, patents, and clinical trials are indispensable types of scientific literature in biomedicine, crucial for knowledge sharing and dissemination. However, these documents are often stored in disparate databases with varying management standards and data formats, making it challenging to form systematic, fine-grained connections among them. To address this issue, we introduce PKG2.0, a comprehensive knowledge graph dataset encompassing over 36 million papers, 1.3 million patents, and 0.48 million clinical trials in the biomedical field. PKG2.0 integrates these previously dispersed resources through various links, including biomedical entities, author networks, citation relationships, and research projects. Fine-grained biomedical entity extraction, high-performance author name disambiguation, and multi-source citation integration have played a crucial role in the construction of the PKG dataset. Additionally, project data from the NIH Exporter enriches the dataset with metadata of NIH-funded projects and their scholarly outputs. Data validation demonstrates that PKG2.0 excels in key tasks such as author disambiguation and biomedical entity recognition. This dataset provides valuable resources for biomedical researchers, bibliometric scholars, and those engaged in literature mining.

Problem

• Connecting disparate biomedical papers, patents, and clinical trials
• Integrating fragmented resources with varying standards and formats
• Enabling systematic knowledge linkages in biomedical research

Innovation

• Integrates papers, patents, and clinical trials
• Links biomedical entities, citations, and projects
• Uses entity extraction and author disambiguation