No abstract

Abstract. This chapter reviews a model of measuring the salience of a specific class of spatial features-façades of buildings-for adaptation to abilities and preferences of user groups of wayfinding services. The model was intentionally designed to be open for such adaptations, and we will report on ways, experiences, and limitations of doing so. We will prove the hypothesis that focalization, i.e., adaptation to different decision situations, can be sufficiently modelled by weights of predetermined salience measures to increase wayfinding success. The long term goal is to identify sets of weights for typical foci of user groups. IntroductionNavigation services calculate routes and present instructions to follow such routes to the user in a geometric format. A typical example might look like this: "Take Street A and go/drive for 150 m; turn left to Street B and go/drive for 70 m; …" Communicating wayfinding information in such a way is not intuitive for the user. Research in spatial cognition has demonstrated that people use landmarks during spatial reasoning and communication of routes (Denis et al. 1999). It is therefore necessary to integrate landmarks into route instructions to enhance their user-friendliness and cognitive plausibility.Recent work has shown a way of automatically extracting salient features from datasets and using these landmarks to enrich wayfinding instructions (Raubal and Winter 2002). The model of landmark saliency is based on three categories of attraction measures, i.e., visual, semantic, and structural. A case study in the city of Vienna was used to demonstrate the applicability and usefulness of the method. Subsequent work included human subject tests to prove the cognitive plausibility of the model (Nothegger et al. 2004).This model described a general approach of measuring the salience of features but it has been applied to a limited case only, i.e., a specific user group (pedestrians in a dense urban environment) travelling in day-light and using specific features (façades of buildings) as landmarks. In principle, the model can be expanded and refined by applying it to different user groups, features, travelling modes, and environments. The goal here is to investigate the fit of the model to different user groups.

No abstract

No abstract

No abstract