Works-magnet: Accelerating Metadata Curation for Open Science

2025-06-17

AI Summary

In open science, bibliographic and research data metadata exhibit high heterogeneity, require complex processing pipelines, and heavily depend on manual curation and validation. Method: This paper proposes an interpretable and editable AI-driven metadata enhancement system. It integrates rule-augmented NLP parsing with institutional affiliation disambiguation algorithms to construct an open-source, auditable processing pipeline. Crucially, it introduces the first visualization of the AI-based metadata generation process, enabling real-time, expert-guided interactive correction. Contribution/Results: The system significantly reduces manual verification effort while improving both accuracy and timeliness of institutional attribution metadata—demonstrated empirically on OpenAlex. Its implementation is fully open-sourced and has been deployed in France’s national open science monitoring infrastructure.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Security and Privacy: Data transparency and provenance
• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

The transition to Open Science necessitates robust and reliable metadata. While national initiatives, such as the French Open Science Monitor, aim to track this evolution using open data, reliance on proprietary databases persists in many places. Open platforms like OpenAlex still require significant human intervention for data accuracy. This paper introduces Works-magnet, a project by the French Ministry of Higher Education and Research (MESR) Data Science&Engineering Team. Works-magnet is designed to accelerate the curation of bibliographic and research data metadata, particularly affiliations, by making automated AI calculations visible and correctable. It addresses challenges related to metadata heterogeneity, complex processing chains, and the need for human curation in a diverse research landscape. The paper details Works-magnet's concepts, and the observed limitations, while outlining future directions for enhancing open metadata quality and reusability. The works-magnet app is open source on github https://github.com/dataesr/works-magnet

Problem

• Accelerating metadata curation for Open Science
• Addressing metadata heterogeneity and complexity
• Reducing human intervention in data accuracy

Innovation

• Automated AI for metadata curation
• Visible and correctable AI calculations
• Open source platform for affiliations