CRII: III: Understanding Urban Vibrancy: A Geographical Learning Approach Employing Big Crowd-Sourced Geo-Tagged Data

点击翻译按钮获取中文摘要

Vibrant communities are defined as places with the following features: permeability, vitality, variety, accessibility, identity, and legibility. Developing vibrant communities can help boost commercial activities, enhance public security, foster social interaction and, thus, yield livable, sustainable, and viable environments. With the advent of the mobile and sensing technologies, big crowd-sourced geo-tagged data (BCGD) are increasingly available from diverse sources (e.g., buildings, vehicles, human, sensors, devices) in urban space, and represent an invaluable source of intelligence for understanding urban vibrancy and enhancing smart growth. This project will develop novel, systematical, and effective analytical techniques to significantly advance critical problems in urban vibrancy by taking advantage of the wealth of BCGD. The algorithms and tools developed in this project will directly impact community planning, city governance, and urban economics. The educational component of this project includes developing a new curriculum that incorporates research into the classroom and provides students from under-represented groups with opportunities to participate in research.This project will develop new analytical techniques to discover, analyze, and leverage the patterns within and the relationships among BCGD to understand and sustain urban vibrancy. Novel methodologies that are appropriate to urban vibrancy will be designed in three directions: measurements, patterns, and mechanism. In researching measurements, axioms that a metric of urban vibrancy should satisfy will be introduced, and principled metrics will be devised to evaluate community vibrancy and learn how it is distributed. In researching patterns, an analytic framework will be proposed to discover the complex patterns of spatial configuration from BCGD for urban vibrancy. This framework aims to identify the compatible dimensions and corresponding measuring methods, as well as optimal portfolios and geographic presentation of spatial configuration. In researching mechanism, new machine learning models will be developed to examine the impact of spatial configuration on urban vibrancy by exploiting the conformity between spatial view and mobility view and the regularity of geographic dependencies.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.