Knowledge Graph Building Workshop

Co-located with the Extended Semantic Web Conference 2019

Portorož, Slovenia - 3 June 2019

See Call for Papers

KGB Call for Papers

More and more Knowledge Graphs are generated for private, e.g. Siri, Alexa, or public use, e.g. DBpedia, Wikidata. While techniques to automatically generate Knowledge Graph from existing Web objects exist (i.e., scraping Web tables), the majority need extensive resources. Those with limited resources typically generate Knowledge Graphs by transforming the content of their datasets (RDB, CSV, etc). Initially, generating Knowledge Graphs from existing datasets was considered an engineering task, however scientific methods recently emerged. Lately, mapping languages for describing rules to generate knowledge graphs and processors to execute those rules emerged. Addressing the challenges related to Knowledge Graphs generation requires well-founded research, including the investigation of concepts and development of tools and methods for their evaluation. KGB is a full-day workshop on Knowledge Graph generation with a special focus on Mapping Languages. It is co-located with the Extended Semantic Web Conference 2019 (ESWC2019). The main goal is to provide a venue for scientific discourse, systematic analysis and rigorous evaluation of languages, techniques and tools, as well as practical and applied experiences and lessons-learnt for generating knowledge graphs from academia and industry.

Topics

  • Mapping based Knowledge Graph Generation
  • Mapping languages for generating Knowledge Graph from legacy datasets
  • Approaches and techniques on (semi)automatically generate mappings
  • Approaches and techniques on collaborative mappings generation
  • Approaches and techniques on exploiting mappings for query answering
  • End User Interfaces (UI) for (collaborative) editing and viewing for Knowledge Graphs building rules and management platforms in general
  • Tools for Knowledge Graph Generation
  • Architectures for Knowledge Graph generation systems
  • (Sustainable) workflows for Web scale Knowledge Graph generation & publishing
  • Methods and Techniques for Knowledge Graph Generation
  • Seamless (distributed) integration/interlinking from heterogeneous data sources
  • Dynamic discovery and retrieval of data for KG generation
  • Quality, Provenance, privacy and trustworthiness of Linked Data generation
  • Knowledge Graph generation and publishing of streaming data at run-time
  • Benchmarks for Knowledge Graphs generation and publishing
  • Lessons learnt, In Use and Experience
  • Experience, lessons learnt and best practices for generating and publishing
  • Negative results and in-use/applied descriptions

Benchmarks

The increasing number of methods and tool for Knowledge Graph construction mandate consensus for evaluation of these methods and tools. We aim to forge this consensus by assessing strengths and weaknesses of tools, increasing the communication among algorithm developers, and improve evaluation techniques and work on Knowledge Graph construction in general. We aim to achieve this by launching an evaluation campaign that includes the publication of test cases and results of different tools’ evaluation.

This year we focus on the test cases that help assessing the coverage of the different tools. We encourage researchers/developers to submit their results after (one of the following):

  • applying the R2RML test cases to their tools that generate knowledge graphs from data in relational databases;
  • applying the RML test cases that we ported from R2RML test cases to their tools that generate knowledge graphs from data in heterogeneous formats;
  • porting the test cases to other languages and applying them to their tools that generate knowledge graphs;
  • applying the RODI benchmark to their tools that (semi-)automatically generates RDB2RDF mappings.

From the results, researchers/developers are expected to submit a system paper (6-8 pages using the LNCS style) in the proceedings of the Knowledge Graph Building workshop.

To ensure easy comparability, we suggest that the papers follow this outline:

  • tools presentation (description, status, algorithms, links to the system);
  • test cases results (comments for the performed test cases and results);
  • discussion and conclusions
    • tool: strengths and weaknesses, possible improvements,
    • test cases: comments on the test cases, new test cases proposal.

Resources:

Authors Guideline

Format

Authors can choose the best way to express their work, such as HTML or PDF. However, a LNCS-like layout should be provided. If your contribution will be in HTML, you can find some available tools in the ESWC19 HTML guideline.

Contributions

  • Full research papers (8-12 pages)
  • In Use and Experience papers (8-12 pages)
  • Position papers (6-8 pages)
  • Short research papers (4-6 pages)
  • System/demo papers (4-6 pages)

Review and Publication

Please, share your contribution before the deadline through the OpenReview platform. The accepted contributions will be published in the proceedings of the workshop. Each accepted paper needs to be presented by one of the authors at the workshop.

Program

Keynote

The invited keynote speaker for the first edition of the KGB workshops is Mariano Rodríguez-Muro, Ontologist in the Knowledge Graph Schema team (Google)

Important dates

11 March, 2019

Submission papers

Submit your paper to KGB workshop using the OpenReview platform. You can see the main topics of the Workshop in the Call for Papers.

29 March, 2019

Notifications

The notification and reviews from our Program Committee will be available.

12 April, 2019

Submission camera ready

Time to have your paper ready for being published. All the accepted paper will be published in the KGB proceedings.

3 June, 2019

Workshop

Keynote, papers presentations, demo jam and a lot of discussion. Remember! If your contribution is accepted, it needs to be presented by one of the authors at the workshop.

Organizers

Anastasia Dimou

Senior Researcher, imec - IDLab (UGent)

David Chaves Fraga

PhD Student, OEG - UPM

Freddy Priyatna

Senior Researcher, OEG - UPM

Juan Sequeda

Capsenta

Pieter Heyvaert

PhD Student, imec - IDLab (UGent)

Program Committee

  • Ahmet Soylu, SINTEF/NTNU
  • Aidan Hogan, Universidad de Chile
  • Amrapali Zaveri, Maastricht University
  • Antoine Zimmermann, École des Mines de Saint-Étienne
  • Ben De Meester, IDLab, Ghent University – imec
  • Boris Villazón-Terrazas, Arvato
  • Claus Stadler, Leipzig University
  • Craig Knoblock, University of Southern California
  • Dumitru Roman, SINTEF/University of Oslo
  • Emanuele Della Valle, Politecnico di Milano
  • Frank Michael, Université Côte d'Azur, CNRS, Inria, I3S
  • Manolis Koubarakis, National & Kapodistrian University of Athens
  • Oscar Corcho, Universidad Politécnica de Madrid
  • Ruben Verborgh, IDLab, Ghent University – imec
  • Soren Auer, Technische Informationsbibliothek

How we built the KGB Web Page

Eating our own dog food

Tools

Most of the information showed on this website comes from the KGB2019 RDF Graph. For generating it, we use the RML suite (YARRRML parser and RML-Mapper) and for querying it, we use the Comunica Framework.

Mappings and CSVs

Using the YARRRML specification, we generate a mapping for building the RDF from the source CSVs. Check the full workflow and see how easy is to generate a Knowledge Graph using mapping languages!