Structural and trend analysis of tagging as a mechanism for organizing learning resources

Abstract

The tagging mechanism has enabled students to organize plenty of online posts, pictures, and academic articles. Previous research has reported that the design of user interfaces for information organization may affect the way of future problem-solving and recall. Over the years, there have been considerable studies on the design of user interfaces for tagging. However, these studies put more emphasis on efficiency of tagging and future recall, but seldom on the structure of tagging outcomes and the tagging process. Therefore, the study designed and compared three types of user interfaces, that is, tag list, tag cloud, and tag network, and then employed network analysis and trend analysis to discover the structural and temporal patterns of tagging. Results of the study showed that, based on the proposed structural indicators and the trend analysis method, the three user interfaces designed for tagging have different degree of effects on tag reusing behaviors. Compared with tag list and tag network, tag cloud can lead more participants to make use of pre-used tags. By visualizing the tagging results as networks for representing the structure of information organization, we further categorized the tagging results into centralized, clustered and separated networks, which can be predicted by the weighted reused times of tags proposed in the study. Besides, participants using tag cloud interface created more clustered networks than those using the other two interfaces. The clustered network structure is corresponding to the selection strategy of the learning resources in the study.