UCTB: An Urban Computing Tool Box for Building Spatiotemporal Prediction Services
Jiangyi Fang, Liyue Chen, Di Chai, Yayao Hong, Xiuhuai Xie, Longbiao Chen, Leye Wang
TL;DR
Spatiotemporal prediction in urban computing faces barriers from integrating diverse domain knowledge and reproducing advanced models. The authors propose UCTB, an open-source toolbox built around a five-step workflow (data conversion, region generation, knowledge management, model definition, training/evaluation) to construct flexible, scenario-general STP services. Key innovations include region-generation to address MAUP, structured knowledge management to fuse temporal, spatial, and external factors, and a library of integrated models with reusable layers, demonstrated on benchmark datasets and a NYC bus ridership use case. Experiments show substantial performance gains through knowledge integration and modular design, suggesting UCTB can accelerate development, improve reproducibility, and support extensible, domain-specific STP services in urban environments.
Abstract
Spatiotemporal crowd flow prediction is one of the key technologies in smart cities. Currently, there are two major pain points that plague related research and practitioners. Firstly, crowd flow is related to multiple domain knowledge factors; however, due to the diversity of application scenarios, it is difficult for subsequent work to make reasonable and comprehensive use of domain knowledge. Secondly, with the development of deep learning technology, the implementation of relevant techniques has become increasingly complex; reproducing advanced models has become a time-consuming and increasingly cumbersome task. To address these issues, we design and implement a spatiotemporal crowd flow prediction toolbox called UCTB (Urban Computing Tool Box), which integrates multiple spatiotemporal domain knowledge and state-of-the-art models simultaneously. The relevant code and supporting documents have been open-sourced at https://github.com/uctb/UCTB.