Figure 4From: Identifying latent activity behaviors and lifestyles using mobility data to describe urban dynamicsRelationship of behavioral patterns with social, mobility, and health problems. (A) Bars show the coefficient of each behavior weight (columns) on the different regression models for each social-economic integration, place exploration, the fraction of people with longer than 45 min commute, amount of distance traveled by day, the fraction of people doing daily physical activity, and the fraction of people not having obesity (rows). (B) Bar shows the \(R^{2}\) for each model, including only the demographic variables D, only the mobility behavioral weights M, and both together \(M+D\). Panel (A) correspond to the coefficients for the \(M+D\) case. See Additional file 1, Supplementary Note 6 for the complete regression tablesBack to article page