Can Social Tags Help You Find What You Want?

One of the uses of social tagging is to associate freely selected terms (tags) to resources for sharing resources among tag consumers. This enables tag consumers to locate new resources through the collective intelligence of other tag creators, and offers a new avenue for resource discovery. This paper investigates the effectiveness of tags as resource descriptors determined through the use of text categorisation using Support Vector Machines. Two text categorisation experiments were done for this research, and tags and web pages from del.icio.us were used. The first study concentrated on the use of terms as its features. The second study used both terms and its tags as part of its feature set. The results indicate that the tags were not always reliable indicators of the resource contents. At the same time, the results from the terms only experiment were better compared to the experiment with terms and tags. A deeper analysis of a sample of tags and documents were also conducted and implications of this research are discussed.