Cell phone call location data has been utilized for the study of travel patterns, but the underlying activities that originate the movement are still at a less explored stage. Resulted from routine and automated features of decision-making processes, human activity and travel behaviour exhibit a high level of spatial-temporal periodicities as well as a certain order of the activities. In this chapter, a method has been developed based on these regularities, which predicts activities being conducted at call locations. The method includes four steps: a set of comprehensive variables is defined; feature selection techniques are applied; a group of state-of-the-art machine learning algorithms and an ensemble of the above algorithms are employed; an additional enhancement algorithm is designed. Using data gathered from natural communication of 80 users over a period of 1 year, the proposed method is evaluated. Based on the ensemble of the models, prediction accuracy of 69.7% was achieved. Using the enhancement algorithm, the performance obtained 7.6% improvement. The experimental results demonstrate the potential to annotate call locations based on the integration between machine learning algorithms and the characteristics of underlying activity and travel behaviour, contributing towards the semantic interpretation and application of the massive data.
Part of the book: Smartphones from an Applied Research Perspective
Activity-based travel demand models predict travel sequences on a day for each individual in a study region. These sequences serve as important input for travel demand estimate and forecast in the area. However, a reliable method to evaluate the generated sequences has been lacking, hampering further development and application of the models. In this chapter, we use travel behavioral information inferred from mobile phone data for such validation purposes. Our method is composed of three major steps. First, locations where a user made calls on a day are extracted from his/her mobile phone records, and these locations form a location trajectory. All the trajectories from the user across multiple days are then transformed into actual travel sequences. The sequences derived from all phone users are further classified into typical patterns which, along with their relative frequencies, define travel profiles. These profiles characterize current travel behavior in the study region and can thus be utilized for assessing sequences generated from activity-based models. By comparing the obtained profiles with statistics drawn from conventional travel surveys, the validation potential of the proposed method is demonstrated.
Part of the book: Mobile Computing