Figure 3From: Identifying latent activity behaviors and lifestyles using mobility data to describe urban dynamicsUsers’ latent behaviors and lifestyles (A) Distribution of weights by latent behavior for three different users in our database. User 1 behavior is dominated by the “Local trips” behavior and has low entropy. However, user 3 has a larger entropy, and her behavior is a very diverse mixture of latent behaviors, including “Bar + Food” and “Coffee Shop”. (B) Average weight for the different latent behaviors in all areas (black) and different cities. (C) Correlation between the weight of latent behaviors and different demographic and urban characteristic features like population density, median income, percentage of Black population, and percentage of people using public transportation. (D) Distribution of user’s weight entropyBack to article page