Organizers 组织者

Abstract / 摘要

• This forum focuses on spatiotemporal data analysis and time series forecasting, aiming to build a high-level interdisciplinary platform for academic exchange. It seeks to bridge traditional statistics with modern artificial intelligence, deeply integrating geoscience, computer science, control theory, and complex systems science. The forum will focus on the intrinsic structure of spatiotemporal dependencies, exploring how to capture dynamic correlations in non-Euclidean spaces and how to leverage deep learning methods such as spatiotemporal graph neural networks, Transformer architectures, Mamba models, and neural differential equations to overcome limitations in traditional time series forecasting.
• The forum particularly emphasizes the integration of data-driven and mechanism-based models, addressing key scientific challenges including non-stationarity, sparsity, multiscale effects, and interpretability in spatiotemporal sequences. It will focus on frontier applications in smart transportation, financial risk control, weather forecasting, and related domains, highlighting how algorithms can be transformed into actionable decision intelligence.

本论坛以“时空数据分析与时序预测”为核心议题，旨在搭建跨学科高端交流平台，融合地理信息科学、计算机科学、控制理论与复杂系统科学的最新进展。 论坛重点关注时空数据依赖性的本质建模，探讨非欧几里得空间中的动态关联刻画方法，并利用时空图神经网络、Transformer结构、Mamba模型以及神经微分方程等先进方法突破传统时序预测瓶颈。 同时强调数据驱动与机理模型融合，重点解决非平稳性、稀疏性、多尺度效应及可解释性等关键科学问题，并面向智慧交通、金融风控、气象预测等典型应用场景，推动算法向决策智能转化。

Topics 主题征稿范围

The following topics are within the scope of this special session, but are not limited to: 以下为本分论坛征稿范围，但不限于：

• Traffic flow prediction and travel demand modeling in smart cities
  智慧城市中的交通流量预测与出行需求建模
• Power load forecasting and renewable energy output prediction
  电力负荷预测与可再生能源出力预测
• Air quality prediction, weather forecasting, and climate modeling
  空气质量预测、气象预报与气候变化建模
• Industrial intelligent systems for fault prediction and supply chain forecasting
  工业智能中的设备故障预测与供应链需求预测
• Financial time series forecasting, market trend analysis, and risk control
  金融经济中的股票价格预测、市场趋势分析与风险控制