Toward equitable,transparent,and collaborative human mobility computing for smart cities

Cities constitute vast and intricate systems in which diverse entities(e.g.,people,vehicles,and roads)interact collaboratively and dynamically.Analyzing and understanding the core elements of people in complex urban mobility systems provides a crucial means for smart city applications.In the era of global digitization,the exponential surge in geolocation data linked to human travel has profoundly transformed our understanding of human travel behavior.Data science enables us to comprehensively capture human mobility characteristics1 at the individual and population levels;these characteristics include regularity,diversity,and predictability.Empowered by data science,human mobility computing research has shaped a closed-loop scientific ecosystem involving data training models,model serving applications,and application feedback data(Figure 1).