Koki Kawabata
Koki Kawabata
川畑 光希
Assistant Professor
SANKEN, The University of Osaka
Email · Google Scholar · GitHub · ORCID
About
I am an Assistant Professor at SANKEN, The University of Osaka. My research focuses on data mining and machine learning for complex, evolving data.
I develop scalable and interpretable methods for data streams, time series, tensors, and dynamic networks, with applications including anomaly detection, event analysis, and forecasting.
Research interests
Data streams and time series
Online modeling, change detection, and forecasting for continuously evolving data.
Tensor and network mining
Scalable methods for multi-aspect data and dynamic interaction structures.
Anomaly detection
Interpretable detection of unusual behavior in complex real-world systems.
Recent publications
RoleMine: Mining Behavioral Patterns in Heterogeneous Call Graphs
Koki Kawabata, Pedro Fidalgo, Mirela Cazalotto, Saranya Vijayakumar, Yasuko Matsubara, Yasushi Sakurai, and Christos Faloutsos. CIKM 2026 (to appear).
Fast Mining and Dynamic Time-to-Event Prediction over Multi-sensor Data Streams
Kota Nakamura, Koki Kawabata, Yasuko Matsubara, and Yasushi Sakurai. KDD 2026.
Interpretable Dynamic Network Modeling of Tensor Time Series via Kronecker Time-Varying Graphical Lasso
Shingo Higashiguchi, Koki Kawabata, Yasuko Matsubara, and Yasushi Sakurai. The Web Conference 2026.