Visual Analytics is the science of analytical reasoning empowered by interactive visualizations. Research in Visual Analytics is closely related to Artificial Intelligence and Machine Learning, since both areas aim to enhance knowledge discovery and decision-making using computational methods such as machine learning, data mining, and modern AI. In contrast to fully automated approaches, Visual Analytics commonly keeps the human in the loop and enables interactive exploration and direct manipulation of underlying models through visual and interactive representations. By leveraging human perception and domain expertise, analysts can reveal patterns, validate hypotheses, detect anomalies, and develop trust in model outcomes that might otherwise remain hidden. With the rise of foundation models, new opportunities emerge to convert unstructured natural language, documents, and LLM-generated outputs into structured representations that can be explored, verified, and refined through interactive visualizations. This tight coupling of extraction, interaction, and visual sensemaking supports transparent, controllable, and auditable AI-assisted analysis workflows. Visual Analytics draws on concepts from computer graphics, information visualization, human-computer interaction, machine learning, artificial intelligence, knowledge discovery, cognition, and visual perception.
The topics of interest include, but are not limited to:
- Combining visual and computational methods for data analysis, machine learning, and artificial intelligence
- Visual Analytics models, pipelines, and interactive approaches
- Human-in-the-loop AI and interactive machine learning
- Visual Trend Analytics and streaming or real-time analytics
- Visual Analytics for spatial, temporal, and spatio-temporal data, including geo-visualization
- Visualization support for multi-criteria decision analysis and decision intelligence
- Knowledge construction, provenance, and sensemaking in Visual Analytics systems
- Guidance, recommendation, and adaptive interaction in Visual Analytics
- Integrative Visual Analytics and AI systems, including hybrid symbolic-neural approaches
- Explainable AI, interpretable machine learning, and uncertainty-aware visualization
- Visual Analytics for model debugging, validation, monitoring, and drift detection
- Visual Analytics for foundation models, including Transformers and Large Language Models (LLMs)
- Visual interaction techniques for prompting, steering, and auditing LLM-based analytics workflows
- Visualization and interaction for embeddings, attention, token-level behavior, and generative outputs
- Retrieval-augmented generation (RAG) and visual interfaces for grounding, citation, and evidence tracing
- Novel techniques to transform LLM outputs into structured representations for interactive visualizations (e.g., entity and relation extraction, schema induction, event extraction, semantic parsing, summarization-to-data, and knowledge graph construction)
- Visual interfaces for exploring, comparing, and validating LLM-generated insights, including interactive citation and evidence tracing
- Human-in-the-loop workflows for correcting and refining extracted structures from LLMs (mixed-initiative extraction, active learning, feedback loops)
- Visual analytics for multi-document synthesis with LLMs (topic structures, argument maps, timelines, and consensus vs. disagreement views)
- Visual Analytics for fairness, accountability, transparency, privacy, and security of AI systems
- Visual analytic solutions for big data challenges, including scalable and distributed approaches
- HCI issues and cognitive approaches for explainable, trustworthy, and accessible Visual Analytics
- Visualization for explaining AI and AI for explaining visualization usage and user intent
- Visualization of machine learning and data mining algorithms and model behavior
- Empirical performance studies, benchmarks, and reproducibility for Visual Analytics and AI systems
- Evaluation methods for visual analytics, visual data mining, and AI-assisted analysis
- Collaborative Visual Analytics and AI-supported collaboration workflows
- Computational steering for long-running machine learning and optimization applications
- Medical Visual Analytics, clinical decision support, and healthcare AI workflows
- Visual Analytics for biomedical data (e.g., EHR, imaging, omics), including clinical pathway analysis
- Intelligent and usable survey technologies, including adaptive questionnaires, conversational surveys, and quality-aware survey analytics
- Reviews and surveys of related literature in Visual Analytics and Artificial Intelligence
Supporting Bodies
Darmstadt University of Applied Sciences, Germany
NOVA LINCS and ISEL-Instituto Politécnico de Lisboa, PT
Organizing Committee
Prof. Kawa Nazemi, Darmstadt University of Applied Sciences, Germany
Prof. Nuno Miguel Soares Datia, NOVA LINCS and ISEL, Instituto Politécnico de Lisboa, PT
Prof. Joao M. Pires, NOVA LINCS Laboratory for Computer Science and Informatics, Universidade NOVA de Lisboa, PT
Dr Loredana Caruccio, University of Salerno, Italy
Dr Cristian A. Secco, Faculty of Computer Science, Darmstadt University of Applied Sciences, Germany
Symposium-specific enquiries should be addressed to the symposium lead organizing coordinators:
Prof. Kawa Nazemi, Darmstadt University of Applied Sciences, Germany
kawa.nazemi (@) h-da.de
Prof. Nuno Miguel Soares Datia, NOVA LINCS and ISEL-Instituto Politécnico de Lisboa, PT
datia (@) isel.ipl.pt