Context

The world of data is changing. Official statistics organizations need to find, acquire and integrate data from both traditional and new data sources at an ever-increasing pace, while maintaining the usability and quality of outputs. Data cycles are shorter, and AI and ML continue to accelerate the need for data.

In this context, the demand for metadata is also changing. Metadata now needs to support FAIR, the Open Science vision and other initiatives that require transparency, reproducibility and data that is not only easy to find and use, but also comparable and interoperable across domains. To this end, we need “smart” metadata that is standard, machine-actionable, capable of enhancing data quality and usability, and rich enough to drive the statistical process.

Smart metadata and AI technologies can benefit from each other: metadata can enable better AI capabilities, but AI also be applied to create and manage better metadata.

Conference Objectives

The COSMOS conference is a place where the official statistics community can work together to define, share, use, and manage smart metadata. Through sharing of experiences, techniques, and tools, and through forward-looking consideration of the challenges we face as a community, we can better meet the demands of the modern data landscape. We encourage official statisticians, metadata experts, data managers, and those involved in the collection, production, and dissemination of statistics to attend.

COSMOS will emphasize the ways in which statistical organizations can implement smart metadata in their institute, covering areas such as:

• Standards
• Business use cases
• Governance
• Best practices definition
• Tools and implementation systems experiences (lessons learned, difficulties encountered, key successes)

This will be the first of a series of conferences, focusing on the best way forward in the development and use of smart metadata in all its forms.

Topics of Interest

Topics of interest include, but not limited to:

Metadata standards

• Lessons learned about implementing standards in your organization
• Any of your own ontologies created for supporting other needs
• Tools used or developed for supporting standards
• Best practices defined
• Interoperability issues between standards
• How to reconcile the conceptual and implementation models?
• How do you define core metadata for your organisation?
• How do you manage metadata (any burden issue?)?
• Governance
• Sharing knowledge about standards in the community
• Feedback: when do you decide to track changes (versioning)?
• How far should we go in terms of traceability? / How far do we version?
• Any experience about provenance and lineage models and implementations
• How to collect various kinds of metadata and standardise them for statistical use?

Linking

• How can data and metadata concepts be linked?
• Linking with geospatial representations
• What are the existing tools?
• How do we establish, document and share best practices?
• How can we perform data reconciliation, ontology matching and instance matching with statistical data?
• Linking codes and concepts between institutes

Active metadata

The use of metadata can extend beyond their function of describing and helping to understand data. Based on machine-actionable semantically rich standard, metadata can automatically generate components of the statistical process and/or trigger their executions. They take on a new role to some extent, no longer simply “information to facilitate understanding of statistics” but becoming “data used in the statistical business process”; hence the notion of active metadata.

Active metadata can serve as an important resource for data management and security, determining quality, providing appropriate access, personalizing delivery of data, and developing flexible “self-serve” systems for better meeting users’ needs.

Experiences of early or mature implementations are welcomed.

FAIR

FAIR data meet the principles of findability, accessibility, interoperability, and reusability.

• How to assess the compliance FAIR principles?
• How to implement FAIR principles in an organization?
• What are FAIR and transparent metadata for confidential data that are shared under restrictions?
• What role do smart metadata play in providing access to protected data?
• What does “reusable” mean for statistical data?
• How does provenance metadata help in being FAIR?

Important Dates

• Submission deadline: September 29 October 13, 2023 (deadline extended)
• Notification of results: October 30 November 13, 2023
• Revised version due: January 15 January 29, 2024

Submission Guidelines

We invite the submission of original results related to the focus areas of the conference in one of the three categories given below:

• Full papers (up to 15 pages)
• Short papers (6-8 pages)
• Posters and demos (up to 4 pages)

All submissions must be written in English and submitted non-anonymously using the EasyChair conference management system at https://easychair.org/conferences/?conf=cosmos2024.

All submitted papers will be subject to a peer-review process and evaluated according to their originality, technical content, style, clarity, and relevance to the conference. At least one co-author of each accepted paper is required to register to COSMOS and present the paper.

The conference proceedings shall be submitted to CEUR-WS.org for online publication. Contribution authors should refer to instructions at https://ceur-ws.org/HOWTOSUBMIT.html to see what this implies. In particular, all contributions must be formatted according to the CEUR-ART style for writing papers to be published with CEUR-WS. Style files and templates are available online: https://ceur-ws.org/Vol-XXX/CEURART.zip. The format adopted by COSMOS 2024 is: 1-column style.

Scientific Committee

• George Alter (University of Michigan)
• InKyung Choi (UNECE)
• Stephane Crête (Statistics Canada)
• Trygve Falch (SSB - Statistics Norway)
• Luis Gonzales (UN Statistical Division)
• Arofan Gregory (CODATA)
• Edgardo Greising (ILO)
• Dan Gillman (Bureau of Labor Statistics)
• Pascal Heus (Postman)
• Simon Hodson (CODATA)
• Matjaž Jug (CBS - Statistics Netherlands)
• Hilde Orten (SIKT)
• Marco Pellegrino (Eurostat)
• Jean-Pierre Poncelet (Eurostat)
• Sónia Quaresma (INE - Statistics Portugal)
• Pascal Rivière (Insee)
• Flavio Rizzolo (Statistics Canada)
• Benoît Rouppert (Meilleurstaux)
• Roxane Silberman (CNRS)

Steering Committee

• Mylène Chaleix (Insee)
• Franck Cotton (Insee)
• Thomas Dubois (Insee)
• Arofan Gregory (CODATA)
• Flavio Rizzolo (Statistics Canada)