Learning Topologies of Situated Public Displays by Observing Implicit User Interactions

In this paper we present a procedure to learn a topological model of Situated Public Displays from data of people traveling between these displays. This model encompasses the distance between different displays in seconds for different ways and/or different travel modes. It also shows how many people travel between displays in each direction. Thus, the model can be used to predict where and when people will appear next after showing up in front of one display. This can be used for example to create continuous ‘shows’ spanning multiple displays while people pass them. To create the model, we use Bluetooth connection data of mobile phones people carry, and employ the EM algorithm to estimate mean travel times for different paths people take.

• Type: Conference
• Year: 2007
• Where: HCI
• Authors: Hans Jörg Müller, Antonio Krüger
• Keywords: Displays, HCI 2007, Human Computer Interaction, Model, Situated Public Displays