Time-Series Forecasting Method Based on Hierarchical Spatio-Temporal Attention Mechanism
In the field of intelligent decision-making, time-series data collected by sensors serves as the core carrier for interaction between the physical and digital worlds. Accurate analysis is the cornerstone of decision-making in critical scenarios, such as industrial monitoring and intelligent transpor...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/13/4001 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849319853209419776 |
|---|---|
| author | Zhiguo Xiao Junli Liu Xinyao Cao Ke Wang Dongni Li Qian Liu |
| author_facet | Zhiguo Xiao Junli Liu Xinyao Cao Ke Wang Dongni Li Qian Liu |
| author_sort | Zhiguo Xiao |
| collection | DOAJ |
| description | In the field of intelligent decision-making, time-series data collected by sensors serves as the core carrier for interaction between the physical and digital worlds. Accurate analysis is the cornerstone of decision-making in critical scenarios, such as industrial monitoring and intelligent transportation. However, the inherent spatio-temporal coupling characteristics and cross-period long-range dependency of sensor data cause traditional time-series prediction methods to face performance bottlenecks in feature decoupling and multi-scale modeling. This study innovatively proposes a Spatio-Temporal Attention-Enhanced Network (TSEBG). Breaking through traditional structural designs, the model employs a Squeeze-and-Excitation Network (SENet) to reconstruct the convolutional layers of the Temporal Convolutional Network (TCN), strengthening the feature expression of key time steps through dynamic channel weight allocation to address the redundancy issue of traditional causal convolutions in local pattern capture. A Bidirectional Gated Recurrent Unit (BiGRU) variant based on a global attention mechanism is designed, leveraging the collaboration between gating units and attention weights to mine cross-period long-distance dependencies and effectively alleviate the gradient disappearance problem of Recurrent Neural Network (RNN-like) models in multi-scale time-series analysis. A hierarchical feature fusion architecture is constructed to achieve multi-dimensional alignment of local spatial and global temporal features. Through residual connections and the dynamic adjustment of attention weights, hierarchical semantic representations are output. Experiments show that TSEBG outperforms current dominant models in time-series single-step prediction tasks in terms of accuracy and performance, with a cross-dataset R<sup>2</sup> standard deviation of only 3.7%, demonstrating excellent generalization stability. It provides a novel theoretical framework for feature decoupling and multi-scale modeling of complex time-series data. |
| format | Article |
| id | doaj-art-5fe5089acf7942e4a0ef737e27275e8c |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-5fe5089acf7942e4a0ef737e27275e8c2025-08-20T03:50:17ZengMDPI AGSensors1424-82202025-06-012513400110.3390/s25134001Time-Series Forecasting Method Based on Hierarchical Spatio-Temporal Attention MechanismZhiguo Xiao0Junli Liu1Xinyao Cao2Ke Wang3Dongni Li4Qian Liu5School of Computer Science & Technology, Beijing Institute of Technology, Beijing 100811, ChinaCollege of Computer Science and Technology, Changchun University, Changchun 130022, ChinaCollege of Computer Science and Technology, Changchun University, Changchun 130022, ChinaCollege of Computer Science and Technology, Changchun University, Changchun 130022, ChinaSchool of Computer Science & Technology, Beijing Institute of Technology, Beijing 100811, ChinaCollege of Computer Science and Technology, Changchun University, Changchun 130022, ChinaIn the field of intelligent decision-making, time-series data collected by sensors serves as the core carrier for interaction between the physical and digital worlds. Accurate analysis is the cornerstone of decision-making in critical scenarios, such as industrial monitoring and intelligent transportation. However, the inherent spatio-temporal coupling characteristics and cross-period long-range dependency of sensor data cause traditional time-series prediction methods to face performance bottlenecks in feature decoupling and multi-scale modeling. This study innovatively proposes a Spatio-Temporal Attention-Enhanced Network (TSEBG). Breaking through traditional structural designs, the model employs a Squeeze-and-Excitation Network (SENet) to reconstruct the convolutional layers of the Temporal Convolutional Network (TCN), strengthening the feature expression of key time steps through dynamic channel weight allocation to address the redundancy issue of traditional causal convolutions in local pattern capture. A Bidirectional Gated Recurrent Unit (BiGRU) variant based on a global attention mechanism is designed, leveraging the collaboration between gating units and attention weights to mine cross-period long-distance dependencies and effectively alleviate the gradient disappearance problem of Recurrent Neural Network (RNN-like) models in multi-scale time-series analysis. A hierarchical feature fusion architecture is constructed to achieve multi-dimensional alignment of local spatial and global temporal features. Through residual connections and the dynamic adjustment of attention weights, hierarchical semantic representations are output. Experiments show that TSEBG outperforms current dominant models in time-series single-step prediction tasks in terms of accuracy and performance, with a cross-dataset R<sup>2</sup> standard deviation of only 3.7%, demonstrating excellent generalization stability. It provides a novel theoretical framework for feature decoupling and multi-scale modeling of complex time-series data.https://www.mdpi.com/1424-8220/25/13/4001time-series forecastingBiGRUSENetGlobalAttention |
| spellingShingle | Zhiguo Xiao Junli Liu Xinyao Cao Ke Wang Dongni Li Qian Liu Time-Series Forecasting Method Based on Hierarchical Spatio-Temporal Attention Mechanism Sensors time-series forecasting BiGRU SENet GlobalAttention |
| title | Time-Series Forecasting Method Based on Hierarchical Spatio-Temporal Attention Mechanism |
| title_full | Time-Series Forecasting Method Based on Hierarchical Spatio-Temporal Attention Mechanism |
| title_fullStr | Time-Series Forecasting Method Based on Hierarchical Spatio-Temporal Attention Mechanism |
| title_full_unstemmed | Time-Series Forecasting Method Based on Hierarchical Spatio-Temporal Attention Mechanism |
| title_short | Time-Series Forecasting Method Based on Hierarchical Spatio-Temporal Attention Mechanism |
| title_sort | time series forecasting method based on hierarchical spatio temporal attention mechanism |
| topic | time-series forecasting BiGRU SENet GlobalAttention |
| url | https://www.mdpi.com/1424-8220/25/13/4001 |
| work_keys_str_mv | AT zhiguoxiao timeseriesforecastingmethodbasedonhierarchicalspatiotemporalattentionmechanism AT junliliu timeseriesforecastingmethodbasedonhierarchicalspatiotemporalattentionmechanism AT xinyaocao timeseriesforecastingmethodbasedonhierarchicalspatiotemporalattentionmechanism AT kewang timeseriesforecastingmethodbasedonhierarchicalspatiotemporalattentionmechanism AT dongnili timeseriesforecastingmethodbasedonhierarchicalspatiotemporalattentionmechanism AT qianliu timeseriesforecastingmethodbasedonhierarchicalspatiotemporalattentionmechanism |