This is my application for the 2020 SWSA Distinguished Dissertation Award regarding my dissertation Storing and Querying Evolving Knowledge Graphs on the Web.

General details

  • Dissertation (HTML / PDF)
  • Short paper (PDF)
  • Supervisor letter (PDF)

Abstract

The Web has become our most valuable tool for sharing information. Currently, this Web is mainly targeted at humans, whereas machines typically have a hard time understanding information on the Web. Using knowledge graphs, this information can be linked in a structured way, so that intelligent agents can act upon this data autonomously. Current knowledge graphs remain however rather static. As there is a lot of value in acting upon evolving knowledge, there is a need for evolving knowledge graphs, and ways to manage them. As such, the goal of this PhD is to allow such evolving knowledge graphs to be stored and queried, taking into account the decentralized nature of the Web where anyone should be able to say anything about anything. Concretely, four challenges related to this goal are investigated: (1) generation of evolving data, (2) storage of evolving data, (3) querying over heterogeneous datasets, and (4) querying evolving data. For each of these challenges, techniques and algorithms have been developed, which prove to be valuable for storing and querying evolving knowledge graphs on the Web. This work therefore brings us closer towards a Web in which both human and machine can act upon evolving knowledge.

Main publications

Below, the main publications related to my dissertation are listed:

All associated publications

The following provides a chronological overview of all my publications to international conferences and scientific journals that I worked on during my PhD. For each of them, I briefly explain their relationship to my dissertation.

2016

  • Ruben Taelman, Ruben Verborgh, Pieter Colpaert, Erik Mannens, Rik Van de Walle. Continuously Updating Query Results over Real-Time Linked Data. Published in proceedings of the 2nd Workshop on Managing the Evolution and Preservation of the Data Web. CEUR-WS, May 2016.
    Workshop paper that received the best paper award at MEPDaW 2016. Because of this, an extended version was published as well: “Continuous Client-Side Query Evaluation over Dynamic Linked Data”.
  • Ruben Taelman. Continuously Self-Updating Query Results over Dynamic Heterogeneous Linked Data. Published in The Semantic Web. Latest Advances and New Domains: 13th International Conference, ESWC 2016, Heraklion, Crete, Greece, May 29 – June 2, 2016.
    PhD Symposium paper in which I outlined the goals of my PhD.
  • Ruben Taelman, Ruben Verborgh, Pieter Colpaert, Erik Mannens, Rik Van de Walle. Moving Real-Time Linked Data Query Evaluation to the Client. Published in proceedings of the 13th Extended Semantic Web Conference: Posters and Demos.
    Accompanying poster of “Continuously Updating Query Results over Real-Time Linked Data”.
  • Ruben Taelman, Ruben Verborgh, Pieter Colpaert, Erik Mannens. Continuous Client-Side Query Evaluation over Dynamic Linked Data. Published in The Semantic Web: ESWC 2016 Satellite Events, Heraklion, Crete, Greece, May 29 – June 2, 2016, Revised Selected Papers.
    Extended version of “Continuously Updating Query Results over Real-Time Linked Data”. Included in this dissertation as Chapter 5
  • Ruben Taelman, Pieter Colpaert, Ruben Verborgh, Pieter Colpaert, Erik Mannens. Multidimensional Interfaces for Selecting Data within Ordinal Ranges. Published in proceedings of the 7th International Workshop on Consuming Linked Data.
    Introduction of an index-based server interface for publishing ordinal data. I worked on this due to the need of a generic interface for exposing data with some kind of order (such as temporal data), which I identified in the article “Continuously Updating Query Results over Real-Time Linked Data”.
  • Ruben Taelman, Pieter Heyvaert, Ruben Verborgh, Erik Mannens. Querying Dynamic Datasources with Continuously Mapped Sensor Data. Published in proceedings of the 15th International Semantic Web Conference: Posters and Demos.
    Demonstration of the system introduced in “Continuously Updating Query Results over Real-Time Linked Data”, when applied to a thermometer that continuously produces raw values.
  • Pieter Heyvaert, Ruben Taelman, Ruben Verborgh, Erik Mannens. Linked Sensor Data Generation using Queryable RML Mappings. Published in proceedings of the 15th International Semantic Web Conference: Posters and Demos.
    Counterpart of the demonstration “Querying Dynamic Datasources with Continuously Mapped Sensor Data” that focuses on explaining the mappings from raw values to RDF.
  • Ruben Taelman, Ruben Verborgh, Erik Mannens. Exposing RDF Archives using Triple Pattern Fragments. Published in proceedings of the 20th International Conference on Knowledge Engineering and Knowledge Management: Posters and Demos.
    Poster paper in which I outlined my future plans to focus on the publication and querying of evolving knowledge graphs through a low-cost server interface.

2017

  • Anastasia Dimou, Pieter Heyvaert, Ruben Taelman, Ruben Verborgh. Modeling, Generating, and Publishing Knowledge as Linked Data. Published in Knowledge Engineering and Knowledge Management: EKAW 2016 Satellite Events, EKM and Drift-an-LOD, Bologna, Italy, November 19–23, 2016, Revised Selected Papers (2017).
    Tutorial in which I presented techniques for publishing and querying knowledge graphs.
  • Ruben Taelman, Ruben Verborgh, Tom De Nies, Erik Mannens. PoDiGG: A Public Transport RDF Dataset Generator. Published in proceedings of the 26th International Conference Companion on World Wide Web (2017).
    Demonstration of the PoDiGG system for which the journal paper “Generating Public Transport Data based on Population Distributions for RDF Benchmarking” was under submission at that point.
  • Ruben Taelman, Miel Vander Sande, Ruben Verborgh, Erik Mannens. Versioned Triple Pattern Fragments: A Low-cost Linked Data Interface Feature for Web Archives. Published in proceedings of the 3rd Workshop on Managing the Evolution and Preservation of the Data Web (2017).
    Introduction of a low-cost server interface for publishing evolving knowledge graphs.
  • Ruben Taelman, Miel Vander Sande, Ruben Verborgh, Erik Mannens. Live Storage and Querying of Versioned Datasets on the Web. Published in proceedings of the 14th Extended Semantic Web Conference: Posters and Demos (2017).
    Demonstration of the interface introduced in “Versioned Triple Pattern Fragments: A Low-cost Linked Data Interface Feature for Web Archives”.
  • Joachim Van Herwegen, Ruben Taelman, Sarven Capadisli, Ruben Verborgh. Describing configurations of software experiments as Linked Data. Published in proceedings of the First Workshop on Enabling Open Semantic Science (SemSci) (2017).
    Introduction of techniques for semantically annotating software experiments. This work offered a basis for the article on “Comunica: a Modular SPARQL Query Engine for the Web” which was in progress by then.
  • Ruben Taelman, Ruben Verborgh. Declaratively Describing Responses of Hypermedia-Driven Web APIs. Published in proceedings of the 9th International Conference on Knowledge Capture (2017).
    Theoretical work on offering techniques for distinguishing between different hypermedia controls. This need was identified when working on “Versioned Triple Pattern Fragments: A Low-cost Linked Data Interface Feature for Web Archives”.

2018

  • Julián Andrés Rojas Meléndez, Brecht Van de Vyvere, Arne Gevaert, Ruben Taelman, Pieter Colpaert, Ruben Verborgh. A Preliminary Open Data Publishing Strategy for Live Data in Flanders. Published in proceedings of the 27th International Conference Companion on World Wide Web.
    Comparison of pull-based and push-based strategies for querying evolving knowledge graphs within client-server environments.
  • Ruben Taelman, Miel Vander Sande, Ruben Verborgh. OSTRICH: Versioned Random-Access Triple Store. Published in proceedings of the 27th International Conference Companion on World Wide Web.
    Demonstration of “Triple Storage for Random-Access Versioned Querying of RDF Archives” that was under submission at that point.
  • Ruben Taelman, Miel Vander Sande, Ruben Verborgh. Components.js: A Semantic Dependency Injection Framework. Published in proceedings of the The Web Conference: Developers Track.
    Introduction of a dependency injection framework that was developed for “Comunica: a Modular SPARQL Query Engine for the Web”.
  • Ruben Taelman, Miel Vander Sande, Ruben Verborgh. Versioned Querying with OSTRICH and Comunica in MOCHA 2018. Published in proceedings of the 5th SemWebEval Challenge at ESWC 2018.
    Experimentation on the combination of the systems from “Comunica: a Modular SPARQL Query Engine for the Web” and “Triple Storage for Random-Access Versioned Querying of RDF Archives”, which both were under submission at that point.
  • Ruben Taelman, Pieter Colpaert, Erik Mannens, Ruben Verborgh. Generating Public Transport Data based on Population Distributions for RDF Benchmarking. Published in Semantic Web Journal.
    Journal publication on PoDiGG. Included in this dissertation as Chapter 2.
  • Ruben Taelman, Miel Vander Sande, Joachim Van Herwegen, Erik Mannens, Ruben Verborgh. Triple Storage for Random-Access Versioned Querying of RDF Archives. Published in Journal of Web Semantics.
    Journal publication on OSTRICH. Included in this dissertation as Chapter 3.
  • Ruben Taelman, Riccardo Tommasini, Joachim Van Herwegen, Miel Vander Sande, Emanuele Della Valle, Ruben Verborgh. On the Semantics of TPF-QS towards Publishing and Querying RDF Streams at Web-scale. Published in proceedings of the 14th International Conference on Semantic Systems.
    Follow-up work on “Continuous Client-Side Query Evaluation over Dynamic Linked Data”, in which more extensive benchmarking was done, and formalisations were introduced.
  • Ruben Taelman, Hideaki Takeda, Miel Vander Sande, Ruben Verborgh. The Fundamentals of Semantic Versioned Querying. Published in proceedings of the 12th International Workshop on Scalable Semantic Web Knowledge Base Systems co-located with 17th International Semantic Web Conference.
    Introduction of formalisations for performing semantic versioned queries. This was identified as a need when working on “Triple Storage for Random-Access Versioned Querying of RDF Archives”.
  • Joachim Van Herwegen, Ruben Taelman, Miel Vander Sande, Ruben Verborgh. Demonstration of Comunica, a Web framework for querying heterogeneous Linked Data interfaces. Published in proceedings of the 17th International Semantic Web Conference: Posters and Demos.
    Demonstration of “Comunica: a Modular SPARQL Query Engine for the Web”.
  • Ruben Taelman, Miel Vander Sande, Ruben Verborgh. GraphQL-LD: Linked Data Querying with GraphQL. Published in proceedings of the 17th International Semantic Web Conference: Posters and Demos.
    Demonstration of a GraphQL-based query language, as a developer-friendly alternative to SPARQL.
  • Ruben Taelman, Joachim Van Herwegen, Miel Vander Sande, Ruben Verborgh. Comunica: a Modular SPARQL Query Engine for the Web. Published in proceedings of the 17th International Semantic Web Conference.
    Journal publication on Comunica. Included in this dissertation as Chapter 4.

2019

  • Brecht Van de Vyvere, Ruben Taelman, Pieter Colpaert, Ruben Verborgh. Using an existing website as a queryable low-cost LOD publishing interface. Published in proceedings of the 16th Extended Semantic Web Conference: Posters and Demos (2019).
    Extension of “Comunica: a Modular SPARQL Query Engine for the Web” to query over semantically annotated paginated websites.
  • Ruben Taelman, Miel Vander Sande, Joachim Van Herwegen, Erik Mannens, Ruben Verborgh. Reflections on: Triple Storage for Random-Access Versioned Querying of RDF Archives. Published in proceedings of the 18th International Semantic Web Conference (2019).
    Conference presentation on “Triple Storage for Random-Access Versioned Querying of RDF Archives”.
  • Miel Vander Sande, Sjors de Valk, Enno Meijers, Ruben Taelman, Herbert Van de Sompel, Ruben Verborgh. Discovering Data Sources in a Distributed Network of Heritage Information. Published in proceedings of the Posters and Demo Track of the 15th International Conference on Semantic Systems (2019).
    Demonstration of an infrastructure to optimize federated querying over multiple sources, making use of “Comunica: a Modular SPARQL Query Engine for the Web”
  • Raf Buyle, Ruben Taelman, Katrien Mostaert, Geroen Joris, Erik Mannens, Ruben Verborgh, Tim Berners-Lee. Streamlining governmental processes by putting citizens in control of their personal data. Published in proceedings of the 6th International Conference on Electronic Governance and Open Society: Challenges in Eurasia (2019).
    Reporting on a proof-of-concept within the Flemish government to use the decentralised Solid ecosystem for handling citizen data.