Exploring Urban Lifestyles Using a Nonparametric Temporal Graphical Model

Abstract

We propose a new unsupervised nonparametric temporal topic model to discover lifestyle patterns from location-based social networks. By relating the textual content, time stamps, and venue categories associated to user check-ins, our framework detects the predominant lifestyle patterns in a given geographic region. The temporal component of our model allows us to analyse the evolution of lifestyle patterns throughout the year. We provide examples of interesting patterns that have been discovered by our model, and we show that our model compares favourably to existing approaches in terms of lifestyle pattern quality and computation time. We also quantitatively show that our model outperforms existing methods in a time stamp prediction task.