B-TGAT: A Bi-directional Temporal Graph Attention Transformer for Clustering Multivariate Spatiotemporal Data

2025-09-16

AI Summary

Addressing challenges in clustering high-dimensional multivariate spatiotemporal climate data—including complex temporal dependencies, dynamically evolving spatial interactions, and non-stationarity—this paper proposes a hybrid U-Net autoencoder architecture integrated with a Bidirectional Temporal Graph Attention Transformer (B-TGAT). Methodologically, ConvLSTM2D layers extract spatiotemporal features; U-Net’s skip connections preserve multi-scale spatial structure; and the B-TGAT module, embedded at the bottleneck layer, enables joint spatiotemporal modeling and adaptive capture of both local and global dependencies. The learned latent representations are both interpretable and highly discriminative. Extensive experiments on three real-world climate datasets demonstrate that the method significantly improves clustering separation, temporal stability, and accurate identification of climatic phase transitions, outperforming state-of-the-art time-series clustering models.

Technology Category

• Machine Learning: Clustering
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Clustering high-dimensional multivariate spatiotemporal climate data is challenging due to complex temporal dependencies, evolving spatial interactions, and non-stationary dynamics. Conventional clustering methods, including recurrent and convolutional models, often struggle to capture both local and global temporal relationships while preserving spatial context. We present a time-distributed hybrid U-Net autoencoder that integrates a Bi-directional Temporal Graph Attention Transformer (B-TGAT) to guide efficient temporal clustering of multidimensional spatiotemporal climate datasets. The encoder and decoder are equipped with ConvLSTM2D modules that extract joint spatial--temporal features by modeling localized dynamics and spatial correlations over time, and skip connections that preserve multiscale spatial details during feature compression and reconstruction. At the bottleneck, B-TGAT integrates graph-based spatial modeling with attention-driven temporal encoding, enabling adaptive weighting of temporal neighbors and capturing both short and long-range dependencies across regions. This architecture produces discriminative latent embeddings optimized for clustering. Experiments on three distinct spatiotemporal climate datasets demonstrate superior cluster separability, temporal stability, and alignment with known climate transitions compared to state-of-the-art baselines. The integration of ConvLSTM2D, U-Net skip connections, and B-TGAT enhances temporal clustering performance while providing interpretable insights into complex spatiotemporal variability, advancing both methodological development and climate science applications.

Problem

• Clustering high-dimensional multivariate spatiotemporal climate data
• Capturing complex temporal dependencies and spatial interactions
• Overcoming limitations of conventional recurrent and convolutional models

Innovation

• Bi-directional Temporal Graph Attention Transformer
• Time-distributed hybrid U-Net autoencoder
• ConvLSTM2D modules with skip connections