Multivariate long-term time-series forecasting tasks aim to discover information-rich features from historical data to predict data for a long time in the future. Multivariate long-term time-series forecasting tasks are commonly used for power forecasting, weather forecasting, traffic forecasting, etc. However, current predictive models for the tasks are still lacking in capturing the relationships between multiple variables and the features of local details. Meanwhile, the existing models suffer from high computational complexity and unsatisfactory accuracy. In this work, we present an effective and efficient method called multiview time-series graph structure representation (MTGSR). MTGSR constructs relationship graphs from three perspectives: time, dimension, and crossing segments to explore complex structural features in multivariate time series. MTGSR not only has better feature extraction capabilities, but also effectively reduces the computational complexity of the model and imp...

Multivariate long-term time-series forecasting tasks are very challenging tasks in many real-world application areas. Recently, researchers focus on designing robust and effective methods, and have made considerable progress. However, there are several issues with existing models that need to be overcome. First, the lack of a relationship structure between multivariate variables needs to be addressed. Second, most models only have a weak ability to capture local dynamic changes across the entire long-term time-series. Third, current models suffer from high computational complexity and unsatisfactory accuracy. To figure out the abovementioned issues, we propose an effective and efficient method called multiview time-series graph structure representation (MTGSR). MTGSR uses GCNs to construct topological relationships in the multivariate time-series from three different perspectives: time, dimension, and crossing segments. Variation trends in different dimensions are extracted through a d...