On the Role of Social Tags in Filtering Interesting Resources from Folksonomies.
Daniela Godoy · 2010
Social tagging systems allow users to easily cre-ate, organize and share collections of resources (e.g. Web pages, research papers, photos, etc.) in a collaborative fashion. The rise in popu-larity of these systems in recent years go along with an rapid increase in the amount of data con-tained in their underlying folksonomies, thereby hindering the user task of discovering interest-ing resources. In this paper the problem of fil-tering resources from social tagging systems ac-cording to individual user interests using purely tagging data is studied. One-class classification is evaluated as a means to learn how to iden-tify relevant information based on positive ex-amples exclusively, since it is assumed that users expressed their interest in resources by annotat-ing them while there is not an straightforward method to collect non-interesting information. The results of using social tags for personal clas-sification are compared with those achieved with traditional information sources about the user in-terests such as the textual content of Web doc-uments. Finding interesting resources based on social tags is an important benefit of exploiting the collective knowledge generated by tagging activities. Experimental evaluation showed that tag-based classification outperformed classifiers learned using the full-text of documents as well as other content-related sources. 1