Hierarchical Joint Graph Learning and Multivariate Time Series Forecasting
Multivariate time series is prevalent in many scientific and industrial domains. Modeling multivariate signals is challenging due to their long-range temporal dependencies and intricate interactions–both direct and indirect. To confront these complexities, we introduce a method of represe...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10286512/ |
| _version_ | 1797633376347226112 |
|---|---|
| author | Juhyeon Kim Hyungeun Lee Seungwon Yu Ung Hwang Wooyeol Jung Kijung Yoon |
| author_facet | Juhyeon Kim Hyungeun Lee Seungwon Yu Ung Hwang Wooyeol Jung Kijung Yoon |
| author_sort | Juhyeon Kim |
| collection | DOAJ |
| description | Multivariate time series is prevalent in many scientific and industrial domains. Modeling multivariate signals is challenging due to their long-range temporal dependencies and intricate interactions–both direct and indirect. To confront these complexities, we introduce a method of representing multivariate signals as nodes in a graph with edges indicating interdependency between them. Specifically, we leverage graph neural networks (GNN) and attention mechanisms to efficiently learn the underlying relationships within the time series data. Moreover, we suggest employing hierarchical signal decompositions running over the graphs to capture multiple spatial dependencies. The effectiveness of our proposed model is evaluated across various real-world benchmark datasets designed for long-term forecasting tasks. The results consistently showcase the superiority of our model, achieving an average 23% reduction in mean squared error (MSE) compared to existing models. |
| first_indexed | 2024-03-11T11:53:13Z |
| format | Article |
| id | doaj.art-cd62aecc9f5642f7b67431383bd424e1 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-11T11:53:13Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-cd62aecc9f5642f7b67431383bd424e12023-11-09T00:00:27ZengIEEEIEEE Access2169-35362023-01-011111838611839410.1109/ACCESS.2023.332504110286512Hierarchical Joint Graph Learning and Multivariate Time Series ForecastingJuhyeon Kim0Hyungeun Lee1Seungwon Yu2Ung Hwang3Wooyeol Jung4Kijung Yoon5https://orcid.org/0000-0001-5574-3299Department of Electronic Engineering, Hanyang University, Seoul, South KoreaDepartment of Electronic Engineering, Hanyang University, Seoul, South KoreaDepartment of Electronic Engineering, Hanyang University, Seoul, South KoreaDepartment of Electronic Engineering, Hanyang University, Seoul, South KoreaDepartment of Electronic Engineering, Hanyang University, Seoul, South KoreaDepartment of Electronic Engineering, Hanyang University, Seoul, South KoreaMultivariate time series is prevalent in many scientific and industrial domains. Modeling multivariate signals is challenging due to their long-range temporal dependencies and intricate interactions–both direct and indirect. To confront these complexities, we introduce a method of representing multivariate signals as nodes in a graph with edges indicating interdependency between them. Specifically, we leverage graph neural networks (GNN) and attention mechanisms to efficiently learn the underlying relationships within the time series data. Moreover, we suggest employing hierarchical signal decompositions running over the graphs to capture multiple spatial dependencies. The effectiveness of our proposed model is evaluated across various real-world benchmark datasets designed for long-term forecasting tasks. The results consistently showcase the superiority of our model, achieving an average 23% reduction in mean squared error (MSE) compared to existing models.https://ieeexplore.ieee.org/document/10286512/Time series analysislong sequence time series forecastgraph neural networkstructure learningself-attention |
| spellingShingle | Juhyeon Kim Hyungeun Lee Seungwon Yu Ung Hwang Wooyeol Jung Kijung Yoon Hierarchical Joint Graph Learning and Multivariate Time Series Forecasting IEEE Access Time series analysis long sequence time series forecast graph neural network structure learning self-attention |
| title | Hierarchical Joint Graph Learning and Multivariate Time Series Forecasting |
| title_full | Hierarchical Joint Graph Learning and Multivariate Time Series Forecasting |
| title_fullStr | Hierarchical Joint Graph Learning and Multivariate Time Series Forecasting |
| title_full_unstemmed | Hierarchical Joint Graph Learning and Multivariate Time Series Forecasting |
| title_short | Hierarchical Joint Graph Learning and Multivariate Time Series Forecasting |
| title_sort | hierarchical joint graph learning and multivariate time series forecasting |
| topic | Time series analysis long sequence time series forecast graph neural network structure learning self-attention |
| url | https://ieeexplore.ieee.org/document/10286512/ |
| work_keys_str_mv | AT juhyeonkim hierarchicaljointgraphlearningandmultivariatetimeseriesforecasting AT hyungeunlee hierarchicaljointgraphlearningandmultivariatetimeseriesforecasting AT seungwonyu hierarchicaljointgraphlearningandmultivariatetimeseriesforecasting AT unghwang hierarchicaljointgraphlearningandmultivariatetimeseriesforecasting AT wooyeoljung hierarchicaljointgraphlearningandmultivariatetimeseriesforecasting AT kijungyoon hierarchicaljointgraphlearningandmultivariatetimeseriesforecasting |