Knowledge Graph based AI reasoning for dynamic data

We provide expertise on Advanced data analytics for time series and other types of data streams coming from industrial processes, energy datasets, mobility data, etc.

Usage of Knowledge Graphs to represent high-level information and to build explainable AI pipelines. Usage of GNN-based models for stream reasoning on heterogeneous and multimodal data.

• Tataroğlu Özbulak, G. A., Manzo, G., Shrestha, Y. R. & Calbimonte, J.-P. (2025). A comprehensive survey of stream reasoning and its integration with knowledge graphs. Knowledge and Information Systems. https://doi.org/10.1007/s10115-025-02589-x

• Tataroğlu Özbulak, G. A., Shrestha, Y. R. & Calbimonte, J.-P. (2025). STKGNN: Scalable Spatio-Temporal Knowledge Graph Reasoning for Activity Recognition. Proc. Of the 34th ACM International Conference on Information and Knowledge Management CIKM, 2853–2862. https://doi.org/10.1145/3746252.3761147

• Tataroğlu Özbulak, G. A., Shrestha, Y. R. & Calbimonte, J.-P. (2025, November). CAST-GNN: Continual Adaptive Learning for Custom Spatio-Temporal Knowledge Graphs via Graph Neural Networks (in press). Proc. Of the IEEE International Conference on Data Mining ICDM.