ABSTRACT
In the modern world, large cities face the challenges of high urban mobility with implications at diverse levels of ecology and economy. To address these issues, urban and transportation planners apply an integrated approach involving diverse areas of computer science. The use of data analytic tools is a common and regular practice among specialists in the areas of urban planning. In particular, the methods and techniques provided by machine learning and information visualization take a predominant role in helping to understand how cities are used by their inhabitants. Scientists and practitioners created diverse solutions to represent information in dynamic and interactive forms, allowing users to explore and analyze urban data faster and more efficiently in comparison to traditional approaches. However, in dynamic environments, the changes in data (e.g. null values, new fields in streamed data, etc.) and user’s manipulations (e.g. map zoom, filter toggling, etc.) can affect the context of the visualization, leading to inefficient graphical representations. In these cases, the visualization should adapt automatically in order to provide coherent and efficient visual representations to the user. This chapter investigates adaptive graphical representation in visualization in the context of urban mobility and exploring different situations of contextual changes. More specifically, adaptive visualization techniques are applied to depict the use of public transportation, which expresses the movement of inhabitants in the city throughout the day.
