Temporal Graph Clustering

Temporal graph clustering (TGC) aims to group nodes in graphs that evolve over time, leveraging both structural and temporal information to improve clustering accuracy and interpretability. Current research emphasizes developing robust ensemble methods, often incorporating deep learning architectures like graph attention networks and autoencoders, to handle the complexity of large-scale, multivariate temporal data. A significant challenge is the limited availability of large, labeled datasets for model evaluation and benchmarking; however, new datasets are emerging to address this gap. The advancements in TGC have broad implications for various fields, including community detection in social networks, anomaly detection in dynamic systems, and hierarchical forecasting in time series analysis.

Papers