Digging Into the Knowledge Graph

Proposal Overview


October 12th, 2017
Wouter Beek (w.g.j.beek@vu.nl)

A global Knowledge Graph emerges

  1. Digitization of artifacts.
  2. Online publication of artifacts.
  3. Unique naming of artifacts.
  4. Relating/linking of artifacts.

A global Knowledge Graph emerges

  1. Digitization of artifacts.
  2. Online publication of artifacts.
  3. Unique naming of artifacts.
  4. Relating/linking of artifacts.

If everybody starts doing this,
a global Knowledge Graph emerges…

Indexing the LOD Cloud

  • A self-organising process of knowledge creation
  • Current indexing approaches are manual.
  • Difficult to grasp what is there, what it means, and what to reuse.
  • Use knowledge classification schemes (BCC, UDC) as reference systems to develop generic principles of indexing.

Objectives

  • Findability of vocabularies.
  • Recommendations for vocabulary reuse.
  • Guidelines for archiving vocabularies.

Research questions (1/3): KOS

  • Which vocabularies are used in the LOD Cloud?
  • What are the features of vocabularies in the LOD Cloud?
  • How can high-level features of the LOD Cloud be described using general schemes of classification?

Research questions (2/3): SW

  • How to index & search for existing vocabularies?
  • How to automatically find candidates for reuse?
  • How to efficiently store and query LOD vocabularies?

Research questions (3/3): Data Curation

  • What are best practices for curating LOD?
  • How to identify ‘endangered’ vocabularies?
  • How to organize provenance and version control for LOD?

Thank you!

Mail: w.g.j.beek@vu.nl
WWW: wouterbeek.com