Overview
This editorial invites researchers and practitioners to participate in a collaborative effort aimed at developing a comprehensive collection of scientific process schemas. These schemas, inspired by schema.org and designed for use with the Open Research Knowledge Graph (ORKG), will focus on experimental and simulation workflows across various scientific disciplines. The resulting schemas will be published openly and will contribute to a manuscript intended for Nature Scientific Data.
Background & Relevance
In the realm of machine learning, the reliance on scientific data from diverse fields such as materials science, biology, and engineering is growing. However, the underlying processes that generate these datasets often remain unstructured and poorly documented in the literature. This lack of standardization complicates the tasks of comparing experiments, reproducing results, and developing machine learning systems that can effectively reason about variations in scientific methodologies. Establishing standardized, machine-actionable descriptions of scientific processes is essential for advancing reliable scientific machine learning systems.
Key Details
- Objective: Create a community library of schemas for scientific processes.
- Publication: Openly published as ORKG templates.
- Target Journal: Manuscript planned for Nature Scientific Data.
- Data Collection Deadline: April 30, 2026.
- Notification of Contributors: By January 31, 2026.
- Participation Link: Register here.
- Broader List of Processes: Available here.
Eligibility & Participation
This initiative welcomes contributions from individuals and small teams who can provide collections of full-text articles (approximately 50 or more) that describe specific experimental or simulation processes in their respective fields. Participants may also offer expert feedback on automatically mined schemas or engage in schema mining themselves. Co-authorship opportunities are available based on the level of involvement.
Submission or Application Guidelines
To participate in this collaborative effort, interested individuals should complete the registration form linked above. Contributors will be selected and notified by the specified date, and the data collection phase will conclude by the end of April 2026.
More Information
The development of standardized scientific process schemas is crucial for enhancing the transparency and trustworthiness of scientific machine learning. By creating a structured library of workflows, this initiative aims to support FAIR (Findable, Accessible, Interoperable, Reusable) data principles, thereby facilitating reproducible benchmarks and improving the capabilities of machine learning models to understand experimental conditions.
Conclusion
The call for collaboration represents a significant opportunity for researchers to contribute to the advancement of machine learning through the establishment of standardized scientific process schemas. Interested parties are encouraged to explore this initiative, register their interest, and help disseminate this important effort within their networks.
Category: CFP & Deadlines
Tags: fair data, scientific workflows, machine learning, schema.org, data standards, reproducibility, experimental processes, orkg, metadata, collaboration, research, data collection
