Comparing Different Layouts of Tag Clouds: Findings on Visual Perception

Abstract

With the emergence of social tagging systems and the possibility for users to extensively annotate web resources and any content enormous amounts of unordered information and user generated metadata circulate the Web. Accordingly a viable visualisation form needs to integrate this unclassified content into meaningful visual representations. We argue that tag clouds can make the grade. We assume that the application of clustering techniques for arranging tags can be a useful method to generate meaningful units within a tag cloud. We think that clustered tag clouds can potentially help to enhance user performance. In this paper we present a description of tag clouds including a theoretical discourse on the strengths and weaknesses of using them in common Web-based contexts. Further recent methods of semantic clustering for visualizing tag clouds are reviewed. Findings from user studies that investigated the visual perception of differently arranged depictions of tags follow. The main objective consists in the exploration of characteristical aspects in perceptual phenomenons and cognitive processes during the interaction with a tag cloud. This clears the way for useful implications on the constitution and design factors of that visualisation form. Finally a new approach is proposed in order to further develop on this concept.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.