Personalizing map content to improve task completion efficiency
Source: International Journal of Geographical Information Science, Volume 24, Number 5, May 2010 , pp. 741-760(20)
Publisher: Taylor and Francis Ltd
Abstract:Significant interaction challenges arise in both developing and using interactive map applications. Users encounter problems of information overload in using interactive maps to complete tasks. This is further exacerbated by device limitations and interaction constraints in increasingly popular mobile platforms. Application developers must then address restrictions related to screen size and limited bandwidth in order to effectively display maps on mobile devices. In order to address issues of user information overload and application efficiency in interactive map applications, we have developed a novel approach for delivering personalized vector maps. Ongoing task interactions between users and maps are monitored and captured implicitly in order to infer individual and group preferences related to specific map feature content. Personalized interactive maps that contain spatial feature content tailored specifically to users' individual preferences are then generated. Our approach addresses spatial information overload by providing only the map information necessary and sufficient to suit user interaction preferences, thus simplifying the completion of tasks performed with interactive maps. In turn, tailoring map content to specific user preferences considerably reduces the size of vector data sets necessary to transmit and render maps on mobile devices. We have developed a geographic information system prototype, MAPPER (MAP PERsonalization), that implements our approach. Experimental evaluations show that the use of personalized maps helps users complete their tasks more efficiently and can reduce information overload.
Document Type: Research Article
Affiliations: 1: Department of Software and Information Systems, University of North Carolina at Charlotte, Charlotte, NC, USA 2: School of Computer Science and Informatics, University College Dublin, Dublin, Ireland
Publication date: 2010-05-01