Understanding the dynamics of urban areas is essential for urban planners and geomarketing specialists. Social media platforms provide the means for such analysis, which has proved to be challenging, due to problem diversity, data volume, veracity and velocity. The scope of this thesis is to design and develop a system which overcome these obstacles and contributes towards an interactive urban analysis. To this end, this thesis presents a monitoring system, which collects data from the social media services “Foursquare” and “Twitter” in near real­time to provide a consistent stream of information for visual analysis. Furthermore, a visual analytics system has been designed and implemented, that offers methods, supported by visualizations and interaction techniques, for the derivation of knowledge to a variety of urban problems. Finally, two case studies from Tokyo, Japan are presented, which confirm the feasibility of the extracted data and the analysis capabilities of the implemented system.