2013 Annual Meeting
Montréal, Québec, Canada | November 1-5, 2013
Olha Buchel, University of Western Ontario
This paper presents a conceptual model for geobrowsing of text collections. This model combines best features and principles from predecessor models that purported to transform geobrowsing to creative visual browsing in geo-semantic spaces. Within the proposed model semantic document collections linked to maps have at least three conceptual spaces: geographic, semantic, and cognitive. All these spaces must be made visible to users by means of representations. Unlike previous models that visualize geographic and semantic spaces separately and often overlook the importance of cognitive spaces, our model blends all these spaces together. This results in place names being the primary objects with attributes derived from the semantic space. Besides representations, visualizations should support interactions with representations to enable users to switch from one set of details to another based on their needs.
The proposed model was evaluated in a small informal study. GeoPubMed visualization prototype was used for communicating and illustrating the conceptual idea of geobrowsing to users and for testing out how the model would work in the context of a real collection – the collection of medical documents in particular. The results of the study reveal that the model was accepted favorably by the study participants. The participants saw the fit of the model for browsing health-related documents and proposed several use scenarios of their own. They also made recommendations for improving the map index, tours in the visualization, and some representations.