Machine-actionable Knowledge for Earth Science

Introductory talk

  • Speaker: Anna-Lena Lorenz
  • Date: 2025-06-12

Key Points

  • Scientific knowledge is still mostly communicated in human-readable formats.
  • Open Research Knowledge Graph (ORKG) is a system that improve FAIR principles in scientific communication.
  • ORKG describes knowledge with concise, machine-actionable Metadata.
  • ORKG presents result values of publications.
  • The ORKG Python package orkg is available to interact with the ORKG API.
  • With the LaTeX package SciKGTeX, structured publication metadata can be integrated in writing workflows.

My notes

  • Public large-scale scholarly datasets are useful reference corpora for benchmarking metadata extraction workflows.
  • The ORKG seems like a global open-source version of the obsidian graph view with result values included.

My Questions

  • The location parameter is not very specific
  • Confidence levels should be clearly defined
  • How do we handle sources (quality, trustworthiness)?
  • Will this lead to (false) equivalency traps?

Advancing FAIR principles in scientific communication through reborn articles

  • Speaker: Lauren Snyder
  • Date: 2025-06-12

Key Points

  • Machines cannot interpret most scientific data
  • But what if we could produce scientific data in a machine-readable form?
  • The Python package dtreg uses schemas for graphics
  • ORKG Reborn yields machine-readable information about the figures and machine environments under which publications were produced.
  • ORKG reborn also yields machine-readable information about single statements in publications.
  • ORKG reborn focuses on the breakdown of scientific publications into single statements and aims to contextualize them with metadata and persistent identifiers such as DOI.

My notes

  • It seems to be an interesting approach to break down scientific publications into single statements.

How to get the most out of ORKG for your discipline

  • Speaker: Anna-Lena Lorenz
  • Date: 2025-06-12

Key Points

  • Each discipline has different needs and requirements
  • We can set up observatories for our discipline
  • Domain experts work with data curators on this observatory
  • This helps for organizing, ensuring quality standard, creation of templates, promoting ORKG and staying in contact with the development team.

ORKG

Website

ORKG reborn

Website Help page

See also: Open Research Knowledge Graph, Knowledge graph, Metadata, ORKG reborn, MOC Open Science Data and Knowledge Graphs