Tag Clouds: How format and categorical structure affect categorization judgment

Abstract

This paper examines how category judgments are influenced by categorical structure and the formatting of tag clouds. Despite the enormous research on categorization, little research has been directed at investigating whether one person can recognize another's categorical structure. A novel approach to measure similarity and categorical structure is proposed. This approach involves the use of latent semantic analyses to compute semantic distances between category exemplars. The empirical domain will be tag clouds, a new development in social computing that provides a particularly useful paradigm for investigating how people identify the categorical structures of others. Three experiments examine how categorical structure and different formatting styles used in tag clouds might affect categorization. Findings reveal that categorization judgments are influenced by categorical structure and tighter structures result in higher accuracy. Format variables such as font size and sorting order were also found to influence accuracy. Future experimental directions are detailed.