Latest updates and achievements from the DMKI Lab
Kashif Rabbani, M. Lissandrini, and K. Hose received the IFIP TC2 Manfred Paul Award for Excellence in Software: Theory and Practice. The award was given for their paper “Extraction of Validating Shapes from Very Large Knowledge Graphs” published in VLDB 2023 (pdf).
Johannes Schrott has joined the team as a PhD student. Welcome!
We have new exciting PhD position available in our team in knowledge-graph driven factuality and explainability. Expected starting time: fall 2025, application deadline: July 31, 2025. More details are available here.
Our work on "Applicability Assessment of Technologies for Predictive and Prescriptive Analytics of Nephrology Big Data" by R. Stojanov, M. Jovanovik, S. Gramatikov, I. Mishkovski, E. Zdravevski, D. Sasanski, Z. Karapancheva, G. Spasovski, I. Vasileska, T. Eftimov, W. Zhuojun, J. Jankowski, D. Trajanov has been published in Proteomics.
Our work on "Smart SPARQL Advisor: Guiding Users in Query Formulation with Performance Prediction" by A. Moharanaj, M. Lissandrini, and K. Hose has been accepted as a demo paper at VLDB 2025.
Our work on "RDFGraphGen: An RDF Graph Generator based on SHACL Shapes" by M. Jovanovik, M. Vecovska, M. Jakubowski, and K. Hose is now available on arxiv.org: preprint.
Our work on "MultiHal: Multilingual Dataset for Knowledge-Graph Grounded Evaluation of LLM Hallucinations" by E. Lavrinovics, R. Biswas, J. Bjerva, K. Hose is now available on arxiv.org: preprint.
Our work on "Towards Generating Synthetic EHR Knowledge Graphs — a Probabilistic Approach" has been accepted at GOBLIN 2025, authors: M. Jovanovik, E. Milenkova, M. Jakubowski, K. Hose
We’re excited to announce the start of our Horizon Europe MSCA DN project ARMADA, focused on conversational data exploration. The project runs from 2025 to 2029.
The paper "FHIR Lens: A Graph-Based Approach to Semantic EHR Exploration" by D. Tomaszuk, A. Smajevic, T. Sagi, and K. Hose has been accepted for publication at CBMS 2025.
Koumudi Ganepola has joined the team as a PhD student. Welcome!
Our work on "PlanRGCN: Predicting SPARQL Query Performance" has been accepted at VLDB 2025, authors: A. Moharanaj, M. Lissandrini, K. Hose
Our work on "The Limits of Graph Samplers for Training Inductive Recommender Systems" (preprint) has been accepted at VLDB 2025, authors: T. E. Jendal, M. Lissandrini, P. Dolog, K. Hose
Together with M. Jovanovik, T. Sagi, and S. El-Sherbiny I attended TARGET‘s annual consortium meeting in Lund.
K. Hose gave a talk on "Knowledge Graphs: Challenges for GDBMS and AI" at EDBT 2025‘s industry event on Next-Generation Data Management Systems.
Martin presented our work on "Fantastic Tables and Where to Find Them: Table Search in Semantic Data Lakes" at EDBT 2025, authors: M. P. Christensen, A. Leventidis, M. Lissandrini, L. Di Rocco, R. J. Miller, K. Hose
Davide and Matteo presented our work on "Towards Reliable Conversational Data Analytics" at EDBT 2025, authors: S. Amer-Yahia, J. Bogojeska, R. Facchinetti, V. Franceschi, A. Gionis, K. Hose, G. Koutrika, R. Kouyos, M. Lissandrini, S. Maniu, K. Mirylenka, D. Mottin, T. Palpanas, M. Rigotti, Y. Velegrakis
K. Hose gave a keynote on "Introduction to Knowledge Graphs" at the GOBLIN‘s 1st plenary meeting in Prague.
Milos Jovanovik has joined the team in Vienna as postdoc. Welcome!
Our work on "Common Foundations of SHACL, ShEx, and PG-Schema" has been accepted at The Web Conference 2025, authors: S. Ahmetaj, I. Boneva, J. Hidders, K. Hose, M. Jakubowski, J.E.L. Gayo, W. Martens, F. Mogavero, F. Murlak, C. Okulmus, A. Polleres, O. Savkovic, M. Simkus, D. Tomaszuk
K. Hose participated in the Dagstuhl Seminar "Trust and Accountability in Knowledge Graph-Based AI for Self Determination"
Our work on "Knowledge Graphs, Large Language Models, and Hallucinations: An NLP Perspective" has been accepted at the Journal of Web Semantics,, authors: E. Lavrinovics, R. Biswas, J. Bjerva, K. Hose
K. Hose had the honor to co-chair the research track at ISWC 2024 and participated in the conference in Baltimore, USA.
Maxime Jakubowski has joined the team in Vienna as a postdoc. Welcome!
K. Hose did a podcast together with Shellie Boudreau talking about knowledge graphs and their role in data engineering, AI, and LLMs, as well as other aspects beyond research results. (youtoube)
K. Hose gave a kexnote on "Knowledge Graphs: ready to power intelligent systems?" at BDA 2024 in Orléans.
K. Hose participated as Austrian MC member in the kickoff event of the COST Action GOBLIN (Global Network on Large-Scale, Cross-domain and Multilingual Open Knowledge Graphs) in Brussels.
K. Hose gave a talk on "Knowledge Graph Challenges with, without, and for AI" at the BIFOLD Colloquium at TUB.
K. Hose participated in Serafeim Papadias' PhD review committee and visit TU Berlin, topic: Mining Highly Dynamic Graphs, supervisors: Volker Markl, Zoi Kaoudi
Our work on "Fantastic Tables and Where to Find Them: Table Search in Semantic Data Lakes" was accepted at EDBT 2025, authors: M. P. Christensen, A. Leventidis, M. Lissandrini, L. Di Rocco, R. J. Miller, K. Hose
K. Hose participated in Thomas Mulder's PhD review committee and visited TU Eindhoven, topic: Optimization of Recursive Queries on Graphs, supervisors: George Fletcher, Nick Yakovets
Our work on "Hypergraphs with Attention on Reviews for Explainable Recommendation" was accepted to ECIR 2024, authors: T. Jendal, TH Le, H. Lauw, M. Lissandrini, P. Dolog, K. Hose
Our work on "How Does Knowledge Evolve in Open Knowledge Graphs?" was accepted to TGDK, authors: A. Polleres, R. Pernisch, A. Bonifati, D. Dell‘Aglio, D. Dobriy, D. Dumbrava, L. Etcheverry, N. Ferranti, K. Hose, E. Jiminez-Ruiz, M. Lissandrini, A. Scherp, R. Tommasini, J. Wachs
Our work on "Tunable Query Optimizer for Web APIs and User Preferences" was accepted at K-CAP 2023, authors: T. Zeimetz, K. Hose, R. Schenkel
Our work on "moduli: A Disaggregated Data Management Architecture for Data-Intensive Workflows" was accepted at ACM SIGWEB Newsletter, authors: P. Ceravolo, T. Cartarci, M. Console, P. Cudré-Mauroux, S. Groppe, K. Hose, J. Pokorny, O. Romero, R. Wrembel
K. Hose gave a keynote on "No Intelligence without Knowledge" at ADBIS 2023 in Barcelona
Our work on "StarBench: Benchmarking RDF-star Triplestores" was accepted at the QuWeDa 2023 workshop at ISWC 2023, authors: G. Abuoda, C. Aebeloe, D. Dell‘Aglio, A. Keen, K. Hose
Our work on "The Need for Better RDF Archiving Benchmarks" was accepted at the MepDAW 2023 workshop icw. ISWC 2023, authors: O. Pelgrin, R. Taelman, L. Galárraga, K. Hose
Our work on "A Gated Architecture for Inductive Recommendation using Knowledge Graphs" was accepted at KARS 2023 workshop at RecSys 2023, authors: T. E. Jendal, M. Lissandrini, P. Dolog, K. Hose
Our work on "GLENDA: Querying RDF Archives with SPARQL" has won the best demo award at ESWC 2023, authors: O. Pelgrin, R. Taelman, L. Galárraga, K. Hose (details)
Our work on "Graph Neural Networks for Metagenomic Binning" was accepted at the CompBio Workshop at ICML 2023, authors: A. Lamurias, A. Tibo, K. Hose, M. Albertsen, T. D. Nielsen
The 1st international workshop on Data Management for Knowledge Graphs (DMKG) was held on May 29 at ESWC 2023, organizers: C. Aebeloe, A. Azzam, O. Hartig, K. Hose (website)
Our work on "Automated Ontology Evaluation: Evaluating Coverage and Correctness using a Domain Corpus" has won the Best Paper Award at the NLP4KGC workshop, authors: A. Zaitoun, T. Sagi, K. Hose
Our work on "Metagenomic Binning using Connectivity-constrained Variational Autoencoders" was accepted at ICML 2023, authors: A. Lamurias, A. Tibo, K. Hose, M. Albertsen, T. D. Nielsen
Our work on "Optimizing SPARQL Queries over Decentralized Knowledge Graphs" was accepted at the Semantic Web Journal, authors: C. Aebeloe, G. Montoya, K. Hose
Our work on "Environmental impact assessment reports in Wikidata and a Wikibase" was accepted at the SemTech4STLD workshop at ESWC 2023, authors: F. Årup Nielsen, I. Lyhne, D. Garigliotti, A. Butzbach, E. R. Boess, K. Hose, L. Kørnøv
Our work on "GLENDA: Querying RDF Archives with SPARQL" was accepted as a demo at ESWC 2023, authors: O. Pelgrin, R. Taelman, L. Galárraga, K. Hose (details)
K. Hose served as PC-cochair for EDBT 2023 together with S. Bhowmick!
Our work on "Patient Event Sequences for Predicting Hospitalization Length of Stay " was accepted at AIME 2023, authors: E. R. Hansen, T. D. Nielsen, T. Mulvad, M. N. Strausholm, T. Sagi, K. Hose
Our work on "SHACTOR: Improving the Quality of Large-Scale Knowledge Graphs with Validating Shapes" was accepted as a demo at SIGMOD 2023, authors: K. Rabbani, M. Lissandrini, K. Hose (details)
Our work on "Diagnosis Prediction over Patient Data using Hierarchical Medical Taxonomies" was accepted at the HEDAI workshop at EDBT 2023, authors: E. R. Hansen, T. Sagi, K. Hose
Our work on "Recommending Tasks based on Search Queries and Missions" was accepted at the Journal of Natural Language Engineering, authors: D. Garigliotti, K. Balog, K. Hose, J. Bjerva