Abstract
Predicting human mobility flows at different spatial scales is challenged by the heterogeneity of individual trajectories and the multi-scale nature of transportation networks. As vast amounts of digital traces of human behaviour become available, an opportunity arises to improve mobility models by integrating into them proxy data on mobility collected by a variety of digital platforms and location-aware services. Here we propose a hybrid model of human mobility that integrates a large-scale publicly available dataset from a popular photo-sharing system with the classical gravity model, under a stacked regression procedure. We validate the performance and generalizability of our approach using two ground-truth datasets on air travel and daily commuting in the United States: using two different cross-validation schemes we show that the hybrid model affords enhanced mobility prediction at both spatial scales.
Highlights
Modelling and understanding human mobility patterns at different spatial scales and aggregation levels - from single individuals to population displacements - is an important research topic because of a vast number of applications, ranging from urban and transportation planning [, ] and resource allocation [, ] to the prediction of migration flows [, ] and epidemic spreading at local, regional, or worldwide level [ – ].In the last few years, a significant research effort has been made to understand human mobility patterns, both in the laws governing individual human trajectories [, ] and collective movements [, ]
Our findings show that the Flickr traces are representative of the real human mobility and that they can be assimilated into a more theoretical model such as the gravity model
3 Discussion In recent years, the analysis of large mobile phone datasets revealed the high predictability of daily individual whereabouts [ – ], that is explained by the extreme regularity of human behaviour
Summary
Modelling and understanding human mobility patterns at different spatial scales and aggregation levels - from single individuals to population displacements - is an important research topic because of a vast number of applications, ranging from urban and transportation planning [ , ] and resource allocation [ , ] to the prediction of migration flows [ , ] and epidemic spreading at local, regional, or worldwide level [ – ].In the last few years, a significant research effort has been made to understand human mobility patterns, both in the laws governing individual human trajectories [ , ] and collective movements [ , , ]. The radiation model considers human movements as diffusion processes that depend on the population distribution over the space, reproducing Stouffer’s theory of intervening opportunities [ ] Both models are static and require some information in order to be adjusted: in the gravity model, parameters are fitted using real mobility data, provided by an independent source; the radiation model, in its original formulation, Beiró et al EPJ Data Science (2016) 5:30 is parameter-free, but it requires accurate knowledge of the spatial population distribution. Both modelling approaches have been extensively tested, showing advantages and limitations. The radiation model offers very good predictions for commuting patterns between US counties using only population data, but its applicability at different spatial scales has been debated since it does not succeed in capturing commuting inside urban or metropolitan areas [ – ] and it has never been used to model long distance travel patterns either
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
