Answers are grounded in this paper and cite their sources.
Accurate temperature prediction plays a key role in various fields such as agriculture, ecology, and the environment. Traditional methods directly learn historical temperature changes to predict future temperature values. However, temperature changes are complex, and traditional methods are difficult to cope with nonlinear changes in temperature, resulting in poor temperature prediction accuracy. To address this issue, we attempt to decompose complex temperature changes into several simple components and propose a multi-frequency temporal convolutional network for short-term temperature prediction. Specifically, three key modules, namely multi - frequency decomposition, component prediction, and parameter optimization are employed to reduce the complexity of temperature changes and the uncertainty of prediction while ensuring that the original temperature information is not lost. On two real temperature datasets, our proposed method outperforms five baselines, which demonstrates that decomposing complex temperature changes is beneficial to improving temperature prediction accuracy.
No takes yet. Share an insight, caveat, or question.
Huang et al. (2024) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: