New York city taxi trips: Dynamic networks following inconsistent power law

摘要

The present paper maps the records of urban taxi trips into dynamic networks, where nodes are the communities and links represent the recorded taxi trips between them. The dynamic urban taxi trip networks, where nodes are the communities and links represent the recorded taxi trips between them, are formulated here as a special type of large-scale traffic system with an enormous impact on the city, in which the existence of uncertainties together with the spatial and temporal variation in the distribution of the taxi trips are considered. Three types of indicators are proposed to facilitate the measurement of the activities between and inside the communities (nodes of the network) from qualitative and quantitative perspectives. It could be found from the analysis of the records within the New York city that these indicators are inconsistent to each other, and nevertheless, none of them distributes uniformly within the city but generally follows the power law in spite of their time-dependent properties. Further, the unusually low values of the scaling parameters from the curve fitting with power law for all the proposed indicators illustrate the severe inhomogeneity of the networks (also the city).