KNowledge on Web Scale

The KNowledge on Web Scale (KNoWS) group of IDLab - Ghent University is an experienced team of creative researchers in advanced data technologies. Our mission involves enabling optimal flows of public and protected data between different parties, and our ultimate goal is for data to impact people’s lives positively in ways they choose themselves. Learn more about KNoWS.

Work with us

Vacancies

Become our colleague and have an impact on the future of the Web! Find out more about available career opportunities and how to apply.

Project

Are you interested in working on a project together with us? Find out more about the possibilities and how to get in touch.

PhD

Are you passionate about the same topics as we are and want to do your PhD with us? Find out more about the possibilities and how to apply.

Research stay

Are you a PhD student who wants to do a research stay with us? Find out more about the possibilities and how to apply.

Master thesis

Are you a student and want to do your master thesis with us? Find out more about the possibilities and how to apply.

Sponsorships

Are you looking for a sponsorship for a course or conference? Do you identify as a women or Black person? Find out more about our sponsorships and how to apply.

ALBATROSS

The ALBATROSS project aims to enhance maritime situational awareness, specifically addressing the emerging threat of Unmanned Aerial Vehicles (UAVs) to critical maritime infrastructure. UAVs have become a significant threat due to advancements in technology, which have increased their range, precision, and payload capacity, while decreasing their footprint and cost. Current detection systems, including radar, camera-based systems, RF sensing, and audio detection, all have limitations in effectively detecting these small, low-flying, and - more and more - autonomous drones. This project seeks to overcome these limitations through the development of an integrated sensor system and intelligent data fusion techniques.

ARCANE

In today’s fast-paced media landscape, editorial teams face an overwhelming challenge: navigating millions of constantly updating media items (articles, social media chatter, etc.) to produce timely, culturally relevant, and fact-based content. Traditional LLMs fall short when handling up-to-date information, niche domains, and culturally sensitive contexts, often hallucinating or failing to provide source attribution. ARCANE tackles this challenge head-on by developing a knowledge-driven agentic AI system that leverages graph-based Retrieval-Augmented Generation (RAG). Unlike conventional RAG systems that rely on shallow text matching, ARCANE builds a Knowledge Graph enriched with cultural context, source metadata, and interlinked content. The AI agent interactively explores this graph, reasoning across long-range connections and maintaining full traceability of sources, ensuring accurate, contextualized, and verifiable outputs. Backed by strong industry demand, supported by partners like VRT, DPG Media, and Roularta, ARCANE aligns EU’s push for trustworthy AI in media and multilingual LLMs. ARCANE is funded by imec.

BOCEMON

BOCEMON is an imec.ICON project, consisting of three industry partners (Quares, DYAMAND and Daikin) and various IDLab teams (PreDICT, KNoWS, HomeLab, Antwerp) within the building management domain. It's a 2-year project that will start 1/10/2024. Project aim: To convert office buildings into truly sustainable, comfortable, healthy and flexible spaces extra services are needed to help building and facility managers to decide on building upgrade investments, optimally allocate office spaces and dynamically control HVAC systems. Such services require access to a broad range of data sources (HVAC data, user feedback, BIM models, etc.) via easy accessible interfaces, but current deployment requires a lot of (manual) work as data is often distributed over many systems, using different protocols and data models. There is a clear need to extend the interoperability capabilities between stakeholders and service providers involved in the building lifecycle, to enable a rapid scaling to many buildings. BOCEMON will develop tools to allow service providers to prototype, test and deploy their services much faster. By using a semantic data model a wide range of building services will be supported and data exchange with diverse stakeholders will be improved. Accompanying tools will be designed by IDLab.KNoWS for easy service design, to explore the available data in the semantic data model, and by IDLab.PreDICT to automatically visualize sensor data on floor plans. IDLab.KNoWS will collaborate with DYAMAND to design an automatic onboarding service to improve the discovery and mapping of sensors and actuators onto this domain model. Hybrid AI based transfer learning methodologies will be designed by IDLab. PreDICT and will allow the efficient transfer of ML models (e.g. to derive trends & patterns on user behavior, occupancy, building health state, etc.) between different building contexts. AI based algorithms will be designed to support investment decisions to realize sustainable and comfortable buildings, and realize a fine-grained occupant-centric and interoperable HVAC control.

CORESPACES

Cities today face growing pressure: rapid urbanisation, climate action targets, rising mobility and energy demand, all while people expect better quality of life. CORESPACES is about helping cities respond to these challenges in a human-centric way. We put EU cities and their people front and center, building local innovation ecosystems where public authorities, industry, academia, and governance actors work together to deploy solutions that are fit for real urban needs, not technology for technology’s sake. At the heart of CORESPACES are the invention of easy-to-use digital tools that combine big data, digital twins, city models, and immersive technologies to help cities: 🔹Rethink how urban spaces are designed 🔹Test decisions before building them 🔹Manage mobility, energy, and public space more efficiently The goal? More inclusive, climate-friendly, resilient, and people-focused cities. These innovations will be tested and validated 9 European Living Labs, including: London, Tartu, Eindhoven/Helmond, Thessaloniki, Limassol, Verona, Cluj-Napoca, Ungheni, and Fribourg, where solutions will be co-designed, tested, and improved in real city settings. Together, we will build blueprints, implementation plans, and policy guidelines enabling innovation solution replications and the ultimate transformation of more EU cities' urban spaces into smarter, inclusive, revitalized, and more resilient environments. The project is funded by the EU and coordinated by Centre for Research & Technology Hellas (CERTH) from Thessaloniki, Greece. Next to imec, 45 multidisciplinary partners make up the project consortium.

DiSHACLed

The DiSHACLed – dissecting data workflows using SHACL – project aims to raise the efficiency of Flemish DSSPs within the European data ecosystem, by starting from the three partners in this consortium: Sirus, Redpencil and Inuits. The general objective of the DiSHACLed project is to increase the efficiency of our data intermediaries within the European data ecosystem. Today each intermediary provides their own tooling within a certain domain. In this project we focus on three categories of hindrances. A first hindrance to data flowing flawlessly is the fact that we are today not able to (i) discover datasets and data services automatically: they need to be manually selected and the URLs to the datasets and their services are hard-coded within applications. This further hinders evolvability as well: as each application is hard-coded against a specific service, it has no mechanism to later switch to another service. Another hindrance is the fact that (ii) tooling from intermediaries – even when this tooling is made available as open-source software – cannot technically be plugged in the tooling of other vendors. It requires manual integration of both technical protocols as the semantics of the domain models. Finally, getting (iii) an overview of the available data, and generating forms for missing data, is at this moment a manual task.

DS4Skills-GO

The DS4Skills-GO project aims to deploy a European Skills Data Space through eight large and sustainable use cases. Building on the infrastructure developed under the EDGE-Skills project, DS4Skills-GO demonstrates the practical benefits of the skills data space, significantly impacting broader EU policies by supporting the European Data Strategy and enhancing digital skills and employment opportunities across Europe. DS4Skills-GO is structured around six key work packages, each focusing on a critical aspect of the project. The project ensures efficient management and quality assurance, establishes robust governance frameworks, maintains data space operations, and focuses on capacity-building, knowledge-sharing, and policy recommendations. Communication, stakeholder engagement, sustainability, and synergy with existing initiatives are also integral parts of the project. The project adopts a "use cases first" approach, leveraging previous work from DS4Skills and EDGE-Skills. It will deploy the skills data space in eight large use cases, addressing challenges like lifelong learning, employee retention, research on learning paths, upskilling, and reskilling. Use case orchestrators will purchase data space services to launch and operate their initiatives. By showcasing the value of the skills data space, DS4Skills-GO aims to enhance digital skills and support the digital transformation of the European economy. The project aligns with the European Data Strategy, emphasising the importance of a thriving data ecosystem while preserving privacy, security, safety, and ethical standards. DS4Skills-GO's efforts are expected to foster innovation, improve education and training policies, and support employment strategies across Europe. DS4Skills-GO addresses immediate skill development needs and contributes to the EU's long-term strategic goals, promoting a skilled and adaptable workforce for the digital age.

GOBLIN

Global Network on Large-Scale, Cross-domain and Multilingual Open Knowledge Graphs. The main aim of the GOBLIN COST Action is to enhance the availability, quality, and utility of public open knowledge across Europe and beyond. Its goal is to develop and support large-scale, high-quality, cross-domain, and multilingual knowledge graphs that are freely usable, reusable, and shareable. We assist in the management committee, and contribute to Working Group 2, primarily on provenance and metadata topics in the context of query execution.

Inclusiversel Tourism

The Inclusiversel Tourism project is an interregional project that aims to address shared barriers to accessible tourism in the regions spanning France, Wallonia, and Flanders. Today, travelers with reduced mobility or specific needs are faced with inconsistent standards and labels, inaccessible infrastructure, and fragmented information, making planning and moving across borders difficult. The goal of the project is to reduce these barriers by connecting, harmonizing, and sharing accessibility data in a clearer and more coherent way, both locally and across the three regions. An AI-based solution will be developed that centralizes and standardizes information about the accessibility of tourist sites. At the same time, an alignment effort is made by the data gathering instances to ensure a consistent experience across borders. With this solution, the project aims to significantly improve the travel experience for people with specific needs and their family or professional caregivers, while also supporting local economic and social development. In addition, AI tools can tailor guidance to user-specific needs instead of generic recommendations, which is expected to support better trip planning through personalized recommendations. By targeting a wide audience—from travelers and caregivers to tourism professionals, local authorities, associations, and academic and research institutions—the project aims to provide tangible value for all partners.

LDT4SSC

The EU project Local Digital Twins for Smart and Sustainable Communities (LDT4SSC) aims to create a robust, scalable, interconnected ecosystem of Local Digital Twins (LDTs), and to galvanise the market for new LDT-based services. This will build on prior achievements, such as the LDT Toolbox and the European data space for smart communities (DS4SSCC-DEP). These efforts will form a resilient LDT infrastructure, enabling European communities to optimise urban services, share resources effectively, and address common challenges like climate change, air quality, waste management, and energy efficiency. By enhancing these infrastructures, communities will be better positioned to use digital twin technology across a range of domains, including urban development, mobility, climate action, and resource management.

MareGraph

MareGraph will build a data architecture for cost-efficient, flexible and sustainable data sharing for a Marine Knowledge Graph. In this project we collaborate with the Flemish Marine Institute (VLIZ), Digital Flanders and the Italian National Research Council (CNR).

PACSOI

PACSOI is a project focused on enabling Solid for healthcare. It’s a two-year project started 1/06/2024. Project aim: Healthcare ecosystems are siloed which makes it difficult to integrate personal health data with clinical data, access and jointly analyze it to generate meaningful insights. These silos are even more challenging for the increasingly common remote monitoring & digital therapeutics tools that gather real-world data & insights. This creates a barrier to leverage these data to achieve a holistic & longitudinal view on the patient’s condition. This is especially troubling for patients with chronic or complex conditions who receive care from multiple providers across settings. Currently, the ownership, control, transparency & consent of health data use & insight generation are outlined in data processing agreements between patients, care providers & third-party companies. This is a complex & time-consuming process, lacking patient involvement in data control decisions, which raises ethical concerns.

SecuWeb

SecuWeb aims to strengthen data security in companies and organisations in Flanders, Wallonia and France. The project investigates how innovative technologies such as Solid, Blockchain, Quantum Internet, 5G, AI and IoT can contribute to safer data usage. Using demonstrators and use cases, we investigate how we can implement these technologies in companies and organisations. We focus on the sectors of e-health, industry 4.0, mobility and food, which are important spearheads for the economy of our regions.

Serendipity Engine

We increasingly rely on algorithmically generated recommendations to navigate in both online and offline contexts: listening to music on streaming platforms, reading news online, or following recommendations about activities and events in your favorite city. These recommender systems help us dealing with the abundance of available information, but at the same time raise questions about their impact on individual citizens and society. The Serendipity Engine project sets out to address these challenges and support societal stakeholders in designing recommender systems to foster serendipity in public contexts.

Solid4Media

With the active involvement of a representative group of Flemish media users, this living lab investigates the potential of Solid technology in a real media context. The goal is to migrate users' media profiles to a pod infrastructure and provide existing and new media services from there. The project proposal focuses on three important use cases identified by Flemish media stakeholders: (i) personal recommendations, (ii) trust through transparency and consent control, and (iii) targeted advertisements. By practically realizing the Solid infrastructure, the technical feasibility and scalability of a future-oriented decentralized media infrastructure is investigated and the concrete added value this can offer for existing and new media services is examined, both from a business and end-user perspective.

SPARK-CS

SPARK-CS stands for Supporting Physical Activity via Recommendations from Knowledge-systems for Cancer Survivors. This project aims to develop a digital health intervention to increase physical activity among cancer survivors through an innovative knowledge-based approach. Despite evidence showing physical activity benefits for cancer survivors, less than 30% meet recommended guidelines. Current digital interventions use rudimentary personalization and contextualisation that fails to address the specific challenges faced by this population. Cancer survivors experience fluctuating symptoms and treatment effects that require highly personalized approaches to physical activity support. SPARK-CS will expand the existing ontologies to create a comprehensive knowledge framework specifically for cancer survivors. The project's novelty lies in its dual-layer decision support system that combines: (1) Ontology-based filtering to ensure recommendations are appropriate & (2) Machine learning algorithms to rank recommendations based on individual factors. Objectives: 1. Expand available ontologies to incorporate cancer-specific concepts and relationships 2. Create an innovative dual-layer decision support system for recommendations 3. Evaluate technical validity and personal relevance of the recommendation in real-world contexts. This project addresses critical gaps in cancer survivorship care through stakeholder co-creation, formal ontology development, and rigorous evaluation strategies. This project is funded by the FWO-KOTK program and combines the expertise of IDLab-UGent-imec and UGent Health Psychology Lab.

TRIPLE

The TRIPLE project, a collaborative effort between the SIB Swiss Institute of Bioinformatics, the University of Ghent and the IOCB Prague, aims to boost the (re)usability of existing knowledge graph resources and improve software tools for RDF data access, documentation and data model visualization. In addition, TRIPLE will increase interoperability between existing public SPARQL endpoints and private data stored in Solid Pods, thus creating an ecosystem of research data that can be seamlessly integrated through efficient and expressive federated SPARQL queries.

W4S

Shipbuilding and maritime supply chain organisations are traditionally based in coastal regions in NWE. Over the last decades coastal regions and the shipping industry alike faced demographic and innovation capacity challenges as especially younger people left these regions and ship(supply) production was largely outsourced to low income countries. As climate change is there to stay, maritime supply chain organisations in coastal regions are challenged to increase their innovation capacity to become fit again to contribute to mitigation solutions. Our W4S approach targets upscaling production and sizes of wing- and rotorsail; in 4 pilot actions we will test and showcase production-, test wingsails and rotorsails on ships and build the body of knowledge for further uptake of Wind Assisted Ship Propulsion (WASP) in the NWE. This will be realized by our excellent partnership that includes, manufacturers, developers of rotor- and wing sails, smart specialization strategists, university and higher education partners and last but not least our launching customers, our shipowners. On the one hand, we will lead WP3, Action 2 to optimize the routes of WAPS vessels based on up-to-date weather data: this allows to optimize the wind propulsion and thus improves the performance of wind-assisted vessels. On the other hand, we will participate in the regional innovations and strategy plan in both WP1 and WP3, focusing on value chain innovation analysis and economics of upscaling for the production of wind-propulsion techniques in the NWEs area. We will also contribute to deliverable outputs, such as data collection and analysis, and scientific publications.

See all projects