Research On Spatio-temporal Population Distributions Based On The Real-world Gathering-scattering Process

The research on population distribution plays a significant role on the research of urban planning,regional economics and many other social fields.This paper gave a detailed study on the spatio-temporal distributions of cityscaled residents by making use of 3G mobile network data and Wi-Fi network data.The inspiration came from the real-world gathering and scattering process of the crowd.We also proposed a spatio-temporal forecasting model of population in a certain area.By including the dynamic mobility information,this model can avoid the effect of data smoothing widely embedded in previous models based on time series or overall spatiotemporal dependence.Through experimental verification,this model has been proved to have a better prediction performance on a fine-grained scale,especially in the prediction of incident crowd.The study and model in this article help to better understand the crowd distribution and provide an effective combination of scientific research and practical application.