Topic Recommendation from Tag Clouds

Abstract

The spread of Web 2.0 has caused user-generated content explosion. Users can tag resources in order to describe and organize them. A tag cloud provides a rough impression of relative importance of each tag within the overall cloud in order to facilitate browsing among numerous tags and resources. The size of a tag cloud may be huge. Thus, the goal of our paper is to recommend topics based on the tag cloud and visualize the recommended topics like a tag cloud. Firstly, an algorithm has been proposed to construct a special tag graph from the tag cloud. Secondly, an algorithm has been provided in order to recommend topics using this tag graph by calculating the reference count of each node. Furthermore, a visualization has been introduced for recommended topics like a tag cloud using a special font distribution algorithm. The proposed graph and algorithms have been validated and verified on the tag cloud of a real-world thesis portal.