arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.01245cs.CL
收藏

SeMob:面向动态城市出行预测的语义合成方法

SeMob: Semantic Synthesis for Dynamic Urban Mobility Prediction

Urban Mobility Institute, Tongji University, China(同济大学城市 mobility 院);College of Surveying and Geo-informatics, Tongji University, China(同济大学测绘地理信息学院);College of Computer Science and Artificial Intelligence, Fudan University, China(复旦大学计算机科学与人工智能学院);Department of Civil Engineering, Toronto Metropolitan University, Canada(多伦多 Metropolitan 大学土木工程系)

Runfei Chen, Shuyang Jiang, Wei Huang

AI总结:

针对现有出行预测模型难利用事件文本应对突发变化的问题,提出基于LLM的SeMob多智能体语义合成框架,通过渐进式融合文本与时空数据提升预测精度,在自建数据集上MAE和RMSE最高分别降低13.92%、11.12%。

摘要:

人类出行预测对城市服务至关重要,但往往无法应对外部事件引发的突发变化。现有时空模型难以利用详细描述这些事件的文本信息。我们提出SeMob,一种基于大语言模型(LLM)的动态出行预测语义合成流水线。具体而言,SeMob采用多智能体框架,其中基于LLM的智能体自动从复杂在线文本中提取并推理时空相关的文本内容。随后,我们提出的创新渐进式融合架构将细粒度相关上下文与时空数据相结合。丰富的预训练事件先验为事件驱动的预测提供了更深入的洞见,从而得到更契合的预测模型。在通过我们的流水线构建的数据集上评估显示,与时空模型相比,SeMob的平均绝对误差(MAE)最多降低13.92%,均方根误差(RMSE)最多降低11.12%。值得注意的是,该框架在事件发生位置和时间附近的时空区域内表现出显著的优越性。

英文摘要:

Human mobility prediction is vital for urban services, but often fails to account for abrupt changes from external events. Existing spatiotemporal models struggle to leverage textual descriptions detailing these events. We propose SeMob, an LLM-powered semantic synthesis pipeline for dynamic mobility prediction. Specifically, SeMob employs a multi-agent framework where LLM-based agents automatically extract and reason about spatiotemporally related text from complex online texts. Fine-grained relevant contexts are then incorporated with spatiotemporal data through our proposed innovative progressive fusion architecture. The rich pre-trained event prior contributes enriched insights about event-driven prediction, and hence results in a more aligned forecasting model. Evaluated on a dataset constructed through our pipeline, SeMob achieves maximal reductions of 13.92% in MAE and 11.12% in RMSE compared to the spatiotemporal model. Notably, the framework exhibits pronounced superiority especially within spatiotemporal regions close to an event's location and time of occurrence.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑