Knowledge Graphs

14 Feb 2025 - 30 Jun 2025

Managing data on one machine for one specific kind of use is fairly straightforward. It is from the moment that that initial dataset needs to be shared with more than one application and needs to be combined with other datasets managed by other organizations on different machines, that more complex computer science and information technology problems arise.

In this course we will deep-dive in the current state of the art in creating Knowledge on WebScale. Your personal data, data published publicly on the Web and data explicitly shared with you, becomes your Knowledge Graph that applications and services can use to assist you in your day to day activities.

Data scientists and engineers today claim 80% of their time goes to preparing and integrating the data: let us take you on a quest to fully automate data integration.

More info & registration ⇗

Practical information:

14 Feb 2025 - 30 Jun 2025

Campus Technologiepark Zwijnaarde – building 131 – room Galileo Ferraris, second floor

English

Target audience: Data scientists and engineers

Want to register?

Prerequisites: A basic knowledge of JavaScript, HTTP, and command line is required
Price: € 329,40
A personal laptop is required

More info & registration ⇗

Georganiseerd door:

Several vacancies in Flanders require knowledge of the Flemish data space in order to share data interoperable. In this series of lessons, the professional learns to prioritize what is important when publishing data, and how he can contribute to a future in which data integration and negotiations over access to data can be fully automated.

Open Data

Web Scraping
Legal aspects of data reuse
Findable, Accessible, Interoperable and Reusable data
Open Data portals

The quest for the universal data model

Knowledge Representations: key–val, resource-based, triple-based
Linked Data and the RDF data model
Linked Data and its serializations
Property graphs and RDF*
Logic with N3

Data Architectures

Linked Data Fragments
Event sourcing and Linked Data Event Streams
RDF Stream Processing
The Open World Assumption
Conway’s law

Web Querying

An introduction to SPARQL
Querying endpoints
Link Traversal
Hypermedia-based querying
Data summaries

Building Linked Data spaces

Data Spaces with IDSA
Metadata management with DCAT
Identity management with Solid-OIDC
Authorization and policies with WAC, ACP, ODRL and N3 rules
Personal data management with Solid
Cross-app interoperability with Solid
Data provenance with PROV-O, P-Plan, SDS
Ontology engineering with SKOS, RDFS and OWL
Validating RDF and building application profiles with SHACL and ShEx

Guest Lectures from European data tech companies and data publishers

Competences

Initial competences

Being able to read HTTP messages (URL, method, body, response codes, headers…)
Executing HTTP requests via the browser and the command-line
Reading and writing data from/in a CSV-file, a JSON-file and relational databases
Making small JavaScript programs in the browser and Node.js (reading files, performing HTTP interactions)

Final competences

Arguing the positioning, importance, and limitations of open data
Choosing the appropriate Web API to publish knowledge graphs
Modeling data as RDF graphs
Publishing knowledge graph on the Web from raw data
Designing a data architecture with fully automated data adoption and assessing trade-offs
Building a Linked Data vocabulary and application profile in RDF
Interpreting and creating SKOS, RDFS, and OWL constraints
Interpreting provenance of RDF data
Performing validation on RDF data
Querying the Web of Linked Data using Comunica
Positioning the industry opportunities and challenges on graph data

Teachers / speakers

Pieter Colpaert

pietercolpaert.be/

Pieter Colpaert is professor aan de Universiteit Gent en verbonden aan imec – IDLab. Hij doet onderzoek naar kostenefficiënte en onderhoudbare Linked Data Web API's. In 2012 was hij medeoprichter van een non-profitorganisatie om Open Knowledge in België te promoten, en vandaag is hij lid van de raad van bestuur van Open Knowledge Belgium. Hij behaalde zijn doctoraat in 2017 aan de Universiteit Gent met de titel “Publishing Transport Data for maximum reuse” waarin hij schetste hoe web-API's kunnen worden ontworpen om de data-integratie volledig te automatiseren, wat hij toepaste op de routeplanning van het openbaar vervoer. Tijdens zijn doctoraat werkte hij echter ook aan andere verticals binnen het e-gov domein, slimme steden, administratieve vereenvoudiging en Open Data in het algemeen. Tijdens zijn postdoc-periode was hij technologiecoördinator van het programma Smart Flanders. In 2021 werd Pieter benoemd tot professor aan de Universiteit Gent in Knowledge on Web-Scale, waar hij Big Data Science en data-architectuur doceerde.

Ruben Verborgh

Ruben Verborgh is a professor of Semantic Web technology at Ghent University – imec and a research affiliate at the Decentralized Information Group at MIT. He aims to build a more intelligent generation of clients for a decentralized Web at the intersection of Linked Data and hypermedia-driven Web APIs. Through the creation of Linked Data Fragments, he introduced a new paradigm for query execution at Web-scale. He has co-authored two books on Linked Data, and contributed to more than 250 publications for international conferences and journals on Web-related topics.