摘要：AbstractDespite the growing availability of big mobility data in cities, methodologies to extract meaningful information from them are still scarce. In this paper, we investigate taxi trips in New York City, develop a large-scale weighted and directed mobility network, and apply a macroscopic methodology to extract the spatial-temporal structure of urban mobility. We also present a new approach to study weighted networks of mobility in which links in the network have journey speed or travel time attribute in addition to commonly used link weights representing number of trips between pairs of nodes. We show that the structure of mobility network in a city when temporal characteristics and variations are taken into account exhibit different properties than what was previously observed. Results provide a better understanding of mobility characteristics in cities.
关键词：KeywordsParsimonious ModelingUrban MobilityTaxiBig DataPower Law
