At the 14th International Joint Conference on Knowledge Graphs (IJCKG 2025), Sophia Sideri presented her work “PG2RDF: Schema-Guided Transformation of Property Graphs to RDF”, offering novel perspectives on bridging property graph and RDF data models. Her presentation focused on novel schema-driven methodologies for transforming property graph data models into RDF representations, highlighting the interoperability challenges and proposing a systematic conversion framework.
Presenter: Sophia Sideri
Paper: PG2RDF: Schema-Guided Transformation of Property Graphs to RDF
Venue: IJCKG 2025 (14th International Joint Conference on Knowledge Graphs)
I’m happy to share that our paper has been presented at ESWC 2025 🎉
This work introduces EntitySum, an approach for entity summarization that leverages centrality measures to identify the most informative and representative aspects of an entity. The goal is to enable more effective exploration and understanding of complex knowledge graphs and semantic data.
Paper: EntitySum: Entity Summarization Using Centrality Measures
Authors: Trouli GE, Vassiliou G, Giannakos I, Papadakis N, Kondylakis H
We are pleased to share that our demo paper has been presented at CIKM 2025 🎉
Paper: White Rabbit: Demonstrating Online KG Pathfinding Using Embeddings
Authors: Antivasis P, Vassiliou G, Tsamis G, Zacharia V, Barka E, Gini J, Kyriakaki A, Andreadakis G, Christodoulakis I, Papadakis N, Kondylakis H
Venue: In Proceedings of the 34th ACM International Conference on Information and Knowledge Management (CIKM ’25)
In this demo, we present White Rabbit, a system for online knowledge graph pathfinding using embeddings, enabling efficient discovery of meaningful paths and relationships between entities. The goal is to support intuitive exploration and improve explainability in graph-based search and analytics.
We are happy to share that our paper has been accepted at EDBT 2026 🎉
Paper: PG-HIVE: Hybrid Incremental Schema Discovery for Property Graphs
Authors: Sideri S, Troullinou S, Ymeralli E, Efthymiou V, Plexousakis D, Kondylakis H
In this work, we present PG-HIVE, a hybrid and incremental approach to schema discovery for property graphs, designed to support evolving graph data and improve interpretability and downstream management tasks.