A collective spatial keyword query is a keyword query that returns geo-textual data in such a way that the result consists of items that cover the keywords collectively, but do not necessarily belong to the same category. The subject of collective spatial keyword querying has seen an increasing scientific interest over the last few years, but efforts have heavily concentrated on the algorithmic side of the problem rather than the visual. This thesis proposes a detailed concept for visualising collective queries on a map and evaluates this concept’s usability and usefulness. A methodology involving all steps from requesting keywords to rendering the results on a map is proposed. Core features are a novel type of collective query called TYPE2a Query, a center-point semantic where the first given keyword denotes a central point for finding matches of the other keywords, typified markers for lower zoom levels and routing between the individual items. To test the hypothesis that collective queries have a higher usability for certain scenarios than current commercial mapping applications without this feature, a prototype called CoSKQVis (Collective Spatial Keyword Query Visualisation) was implemented and a user test was conducted. Participants were randomly divided between two groups and had to solve a task that involved finding a satisfactory set of five points of interest (PoIs) in an unknown environment with and without the feature of collective querying. They were then asked for their experience regarding different aspects of usability like efficiency, learnability, ease of use, subjective satisfaction and accuracy. The results support the hypothesis insofar that the prototype was rated significantly better in several of those aspects than the commercial alternative in the given target group and for the given task. Also usefulness was confirmed by the users for specific scenarios. However, many users in both groups had problems with executing the task entirely correctly and finding a set of five PoIs. It has been shown that there is a need and interest for research in this matter. This thesis introduces the subject of visualising collective keyword queries, paving the way for future research to create an impact on search efficiency in the long term.     «

A collective spatial keyword query is a keyword query that returns geo-textual data in such a way that the result consists of items that cover the keywords collectively, but do not necessarily belong to the same category. The subject of collective spatial keyword querying has seen an increasing scientific interest over the last few years, but efforts have heavily concentrated on the algorithmic side of the problem rather than the visual. This thesis proposes a detailed concept for visualising...     »