© 2014 Springer International Publishing Switzerland. One of the main challenges in urban computing is to interpret the behaviors of the individuals, and so to provide services for suppling their needs. Data mining offers very powerful tools that can be used to analyze data in urban environments. Our research uses the UrbanContext roles model to identify the states of the individuals within urban environments. Then, it applies supervised classification data mining techniques to the results obtained, and uses decision trees in order to facilitate the analysis of the individuals’ behavior. Finally, we present the prediction results obtained from a study made about the roles that individuals adopt depending on their context. From these data we successfully predict the different types of services we can offer in an urban environment.