• Submissions open: June 9, 2025
• Submissions close: October 13, 2025
• First-round review decisions: January 19, 2026
• Deadline for revised submissions: April 13, 2026
• Notification of final decisions: June 8, 2026

Submissions that are received before the first deadline will be directly sent out for review.

We invite submissions on a range of topics:

• Novel algorithms for scalable graph mining and analysis.
• Advances in graph embeddings and graph representation learning (e.g., GNNs).
• Efficient processing of large-scale, heterogeneous, and dynamic graphs.
• Integration of temporal and spatial information in graph models.
• Graph kernels, summarization/coarsening, alignment.
• Graph language, generative, and foundation models.

Evaluation and Benchmarks

• New metrics and benchmarks for graph mining and learning methods.
• Empirical evaluations of graph-based systems in real-world scenarios.
• Analysis of robustness and reliability in graph-based decision systems.

Applications of Knowledge Discovery from Graphs

• Real-world case studies in social media analysis (e.g., misinformation propagation), recommender systems, and computer vision.
• Knowledge graph construction and its use in information retrieval and natural language processing (e.g., retrieval augmented generation with graphs).
• Applications in financial security (e.g., fraud detection), cybersecurity (e.g., malware detection/propagation), and graph ML platforms (e.g., in-database machine learning).
• Use of graph-based techniques in bioinformatics (e.g., drug discovery), transportation/mobility networks (e.g., traffic prediction), and climate science (e.g., global weather forecasting).

Beyond Accuracy in Knowledge Discovery from Graphs

• Interpretable and explainable graph-based methodologies.
• Robustness and adversarial attacks on graphs.
• Responsible AI (e.g., fairness, bias) on graph neural networks.
• Generalization of graph-based approaches on unseen nodes and graph structures.

Emerging Trends and Interdisciplinary Approaches

• Fusion of graph learning with other machine learning paradigms (e.g., federated learning, reinforcement learning).
• Use of knowledge discovery techniques in dynamic and evolving graphs.
• Cross-disciplinary approaches combining KDG with fields like neuroscience, urban planning, and environmental science.

We welcome original research papers, case studies, and review articles that contribute to the body of knowledge in these areas.

• Ludovico Boratto, University of Cagliari (Italy), ludovico.boratto@acm.org
• Daniele Malitesta, Université Paris-Saclay, CentraleSupélec, Inria (France), daniele.malitesta@centralesupelec.fr
• Fragkiskos D. Malliaros, Université Paris-Saclay, CentraleSupélec, Inria (France), fragkiskos.malliaros@centralesupelec.fr
• Mirko Marras, University of Cagliari (Italy), mirko.marras@acm.org
• Giacomo Medda, University of Cagliari (Italy), giacomo.medda@unica.it
• Cataldo Musto, University of Bari Aldo Moro (Italy), cataldo.musto@uniba.it
• Erasmo Purificato, European Commission, Joint Research Centre (Italy), erasmo.purificato@acm.org