CrossLag：基于领域知识的变换器预测大流行性登革热暴发
机器学习
2025-10-07 v1 计算机与社会
摘要
尽管已开发了多种模型来预测登革热病例，但要预测需要及时公共警报的重大登革热暴发仍然具有挑战性。本文引入 CrossLag，通过环境感知注意力机制，在参数计数较少的情况下将滞后内生信号纳入变换器架构。暴发事件通常滞后于气候和海洋异常的重大变化。我们采用最近提出的通用变换器 TimeXer 作为本研究的基线。我们提出的模型在 Singapore dengue data 24 周预测窗口上显著优于 TimeXer，在检测和预测重大暴发方面表现更佳。
引用
@article{arxiv.2510.03566,
title = {CrossLag: Predicting Major Dengue Outbreaks with a Domain Knowledge Informed Transformer},
author = {Ashwin Prabu and Nhat Thanh Tran and Guofa Zhou and Jack Xin},
journal= {arXiv preprint arXiv:2510.03566},
year = {2025}
}
备注
(C) 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works