Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

zbMATH Open: API Solutions and Research Challenges

Matteo Petrera, Dennis Trautwein, Isabel Beckenbach, Dariush Ehsani, Fabian Mueller, Olaf Teschke, Bela Gipp, Moritz Schubotz

TL;DR

zbMATH Open introduces the Scholix-based zbMATH Links API and demonstrates it with the DLMF partnership, enabling open, interoperable bibliographic linking for mathematics. The paper details API structure, data modeling, and current limitations, and outlines plans to onboard MathOverflow, arXiv, and OEIS, expanding cross-platform connectivity. It catalogs immediate research opportunities—ranging from tagging and PDF benchmarking to formula search and disambiguation—and provides a roadmap for AI-assisted analysis of mathematical literature. The work aims to accelerate discovery and reproducibility by providing open, license-friendly access to bibliographic metadata and interlibrary links, and it lays out a concrete set of future enhancements and research questions.

Abstract

We present zbMATH Open, the most comprehensive collection of reviews and bibliographic metadata of scholarly literature in mathematics. Besides our website https://zbMATH.org which is openly accessible since the beginning of this year, we provide API endpoints to offer our data. The API improves interoperability with others, i.e., digital libraries, and allows using our data for research purposes. In this article, we (1) illustrate the current and future overview of the services offered by zbMATH; (2) present the initial version of the zbMATH links API; (3) analyze potentials and limitations of the links API based on the example of the NIST Digital Library of Mathematical Functions; (4) and finally, present the zbMATH Open dataset as a research resource and discuss connected open research problems.

zbMATH Open: API Solutions and Research Challenges

TL;DR

zbMATH Open introduces the Scholix-based zbMATH Links API and demonstrates it with the DLMF partnership, enabling open, interoperable bibliographic linking for mathematics. The paper details API structure, data modeling, and current limitations, and outlines plans to onboard MathOverflow, arXiv, and OEIS, expanding cross-platform connectivity. It catalogs immediate research opportunities—ranging from tagging and PDF benchmarking to formula search and disambiguation—and provides a roadmap for AI-assisted analysis of mathematical literature. The work aims to accelerate discovery and reproducibility by providing open, license-friendly access to bibliographic metadata and interlibrary links, and it lays out a concrete set of future enhancements and research questions.

Abstract

We present zbMATH Open, the most comprehensive collection of reviews and bibliographic metadata of scholarly literature in mathematics. Besides our website https://zbMATH.org which is openly accessible since the beginning of this year, we provide API endpoints to offer our data. The API improves interoperability with others, i.e., digital libraries, and allows using our data for research purposes. In this article, we (1) illustrate the current and future overview of the services offered by zbMATH; (2) present the initial version of the zbMATH links API; (3) analyze potentials and limitations of the links API based on the example of the NIST Digital Library of Mathematical Functions; (4) and finally, present the zbMATH Open dataset as a research resource and discuss connected open research problems.

Paper Structure

This paper contains 23 sections, 6 figures, 1 table.

Table of Contents

  1. Introduction
  2. DLMF as a zbMATH partner
  3. zbMATH Links API
  4. Structure of the API
  5. Analysis of DLMF Data
  6. Usage
  7. Limitations and Future Partners
  8. Research Opportunities
  9. Immediate Research Opportunities
  10. Tagging of Scientific Publications
  11. PDF Text Extraction Benchmark
  12. Training Dataset
  13. Formula Search
  14. Recommender Systems
  15. Formula Disambiguation I
  16. ...and 8 more sections

Figures (6)

  • Figure 1: Conceptual overview of the zbMATH database and its associated ingress and egress data flows to various components. This paper puts the emphasis on the "Scholix Links API".
  • Figure 2: A reference in DLMF, available at https://dlmf.nist.gov/bib/O (below), and a link to it, https://dlmf.nist.gov/2.10#iv.p2 (above)
  • Figure 3: Number of links to the zbMATH API. One can see a huge increase in 2010 -- the year DLMF officially started.
  • Figure 4: Distribution of primary 2-digit MSC codes in the DLMF dataset
  • Figure 5: Distribution of years of publication of references in the DLMF dataset
  • ...and 1 more figures