Data for Integrating CyberGIS and Urban Sensing for Reproducible Streaming Analytics

Increasingly pervasive location-aware sensors interconnected with rapidly advancing wireless network services are motivating the development of near-real-time urban analytics. This development has revealed both tremendous challenges and opportunities for scientific innovation and discovery. However, state-of-the-art urban discovery and innovation are not well equipped to resolve the challenges of such analytics, which in turn limits new research questions from being asked and answered. Specifically, commonly used urban analytics capabilities are typically designed to handle, process, and analyze static datasets that can be treated as map layers and are consequently ill-equipped in (a) resolving the volume and velocity of urban big data; (b) meeting the computing requirements for processing, analyzing, and visualizing these datasets; and (c) providing concurrent online access to such analytics. To tackle these challenges, we have developed a novel cyberGIS framework that includes computationally reproducible approaches to streaming urban analytics. This framework is based on CyberGIS-Jupyter, through integration of cyberGIS and real-time urban sensing, for achieving capabilities that have previously been unavailable toward helping cities solve challenging urban informatics problems. The files included in this dataset functions as follows: 1) Spatial_interpolation.ipynb is a python based Jupyter notebook that enables users to conduct spatial interpolation with AoT data; 2) Urban_Informatics.ipynb is a Jupyter notebook that helps to explore the AoT dataset; 3) chicago-complete.weekly.2019-09-30-to-2019-10-06.tar includes all the high-frequency urban sensing data from AoT sensors from 2019 September 30th to 2019 October 6th collected in Chicago, US; 4) sensors.csv is a processed dataset including information about the temperature in Chicago, and it is used in Spatial_interpolation.ipynb.

随着日益普及的位置感知传感器与快速演进的无线网络服务深度互联，近实时城市分析的研发工作正迎来强劲驱动力。这一发展趋势既为科学创新与发现带来了巨大机遇，也暴露了诸多严峻挑战。然而，当前最先进的城市发现与创新研究框架，尚不足以应对此类分析任务面临的挑战，进而制约了新型研究问题的提出与解答。具体而言，主流城市分析工具通常仅针对可作为地图图层的静态数据集设计开发，因此在以下三方面存在明显局限：(a) 难以适配城市大数据的体量与实时流速；(b) 无法满足此类数据集处理、分析与可视化的计算需求；(c) 难以支持此类分析任务的并发在线访问。为应对上述挑战，本团队研发了一款新型网络地理信息系统（cyberGIS）框架，其中包含支持流式城市分析的可计算复现方法。该框架基于CyberGIS-Jupyter，通过整合网络地理信息系统与实时城市感知技术，实现了此前无法达成的功能，可助力城市解决复杂的城市信息学难题。本数据集包含的文件功能说明如下：1) Spatial_interpolation.ipynb：基于Python的Jupyter Notebook，支持用户利用AoT数据开展空间插值分析；2) Urban_Informatics.ipynb：Jupyter Notebook，可辅助用户探索AoT数据集；3) chicago-complete.weekly.2019-09-30-to-2019-10-06.tar：包含2019年9月30日至2019年10月6日期间，美国芝加哥市由AoT传感器采集的全部高频城市感知数据；4) sensors.csv：经过预处理的数据集，包含芝加哥市气温相关信息，可用于Spatial_interpolation.ipynb的分析流程。