Description

This tutorial introduces participants to the X3ML Framework, a suite of specifications and tools for constructing knowledge graphs from structured data. The tutorial will provide both conceptual foundations and hands-on demonstrations, guiding attendees through the workflow of defining schema mappings and transforming source data into RDF aligned with domain ontologies. At the core of the framework is a declarative, technology-agnostic mapping language that supports collaborative and maintainable schema mapping practices. Participants will learn how X3ML facilitates not only the creation of semantic mappings but also the exploration and verification of the resulting knowledge graphs to ensure correctness and completeness. The session will further highlight domain-independent applications of X3ML across fields such as cultural heritage and biodiversity, while also showcasing emerging methods that integrate Large Language Models (LLMs) to accelerate and simplify schema mapping tasks.

Learning Objectives

By the end of this tutorial participants will be able to:

  • understand the principles and challenges of transforming heterogenous source data into RDF knowledge graphs
  • apply the X3ML mapping language and associated tools to design, implement and validate schema mappings
  • gain hands-on experience in generating semantically rich knowledge graphs and verifying their quality
  • explore innovative techniques leveraging Large Language Models (LLMs) to reduce manual effort and enhance the schema mapping process
Equipped with this knowledge, participants will be able to effectively use X3ML framework in real-world scenarios to build interoperable, semantically robust knowledge graphs across diverse application domains.

Format and Schedule

The tutorial will be held on October 16, 2025 as part of the IJCKG 2025 conference, and it will be structured to balance conceptual understanding with practical application. It will begin with a slide-based presentation introducing the core concepts of knowledge graph construction and the X3ML framework. This will be followed by a live online demonstration, showcasing the workflow of schema mapping and RDF generation using the framework tools and resources. Finally, participants will engage in guided hands-on exercises to practice defining schema mappings and transforming data collections themselves. The entire session will last one hour, ensuring an engaging and focused learning experience that combines theory, demonstration and active participation within the conference program.

Audience

This tutorial is designed for researchers, practitioners, and developers interested in semantic technologies and data integration for the construction of knowledge graphs. It will be particularly valuable for those working with ontologies and knowledge graphs across domains such as cultural heritage, and other data-intensive fields. Participants are expected to have a basic understanding of ontologies and RDF, which will help them follow the conceptual discussions and fully benefit from the practical hands-on exercises. No prior experience with the X3ML framework is required, as all necessary concepts and tools will be introduced during the session.

Presenter

The tutorial will be presented by Yannis Marketakis. Yannis is a R&D software engineer and technical manager at the Information Systems Laboratory and the Centre for Cultural Informatics of FORTH-ICS. He is also a PhD candidate and holds an MSc in Information Systems. His research interests include information systems, digital preservation, conceptual modelling, knowledge representation, semantic data integration, and software engineering for large-scale infrastructures. He has a vast experience in conducting applied research in the field of ICT. He has been actively involved in several European and national research projects contributing as lead R&D engineer. He is co-author of more than 55 scientific publications and one book, and has presented his work at various international conferences. His contributions have been recognized with a Best Paper Award (MTSR'20). He brings deep expertise in the design and implementation of semantic data integration workflows for the construction of knowledge graphs.