Mining Association Rules in Folksonomies Christoph Schmitz, Andreas Hotho, Robert Jaschke,2, Gerd Stumme, 2 1 Knowledge Data Engineering Group, Department of Mathematics and Computer Science, University of Kassel, wilhelmshher Allee 73. D-34121 Kassel Germanyhttp://www.kde.cs.uni-kassel.de 2 Research Center L3S, Expo Plaza 1, D-30539 Hannover, Germany, Abstract. Social bookmark tools are rapidly emerging on the Web. In such syster users are setting up lightweight conceptual structures called folksonomies. These sys- tems provide currently relatively few structure. We discuss in this paper, how assc ciation rule mining can be adopted to analyze and structure folksonomies, and how the results can be used for ontology learning and supporting emergent semantics. We demonstrate our approach on a large scale dataset stemming from an online system 1 Introduction A new family of so-called Web 2.0applications is currently emerging platforms like Wikis, Blogs, and social resource sharing systems. In this paper e the Web. These include user-centric publishing and knowledge manageme we focus on resource sharing systems, which all use the same kind of lightweight knowledge representation, called folksonomy. The word'folksonomy'is a blend of the words 'taxonomy'andfolk', and stands for conceptual structures created by the people Resource sharing systems, such as Flickr or del icio us, have acquired large numbers of users(from discussions on the del icio us mailing list, one can ap- proximate the number of users on del icio us to be more than one hundred thousand) within less than two years. The reason for their immediate success is the fact that no specific skills are needed for participating, and that these tools yield immediate benefit for each individual user (e. g. organizing ones bookmarks in a browser-independent, persistent fashion) without too much overhead. Large numbers of users have created huge amounts of information within a very short period of time. As these systems grow larger, however, the users feel the need for more structure for better organizing their resources. For instance, approaches for tagging tags, or for bundling them, are currently dis- cussed on the corresponding news groups. Currently, however, there is a lack of theoretical foundations adapted to the new opportunities which has to be A first step towards more structure within such systems is to discover knowl- edge that is already implicitly present by the way different users assign tags to resources. This knowledge may be used for recommending both a hierarchy Ihttp://www.flickr.com http://del.icio.usMining Association Rules in Folksonomies Christoph Schmitz1 , Andreas Hotho1 , Robert J¨aschke1,2 , Gerd Stumme1,2 1 Knowledge & Data Engineering Group, Department of Mathematics and Computer Science, University of Kassel, Wilhelmshher Allee 73, D–34121 Kassel, Germany, http://www.kde.cs.uni-kassel.de 2 Research Center L3S, Expo Plaza 1, D–30539 Hannover, Germany, http://www.l3s.de Abstract. Social bookmark tools are rapidly emerging on the Web. In such systems users are setting up lightweight conceptual structures called folksonomies. These sys￾tems provide currently relatively few structure. We discuss in this paper, how asso￾ciation rule mining can be adopted to analyze and structure folksonomies, and how the results can be used for ontology learning and supporting emergent semantics. We demonstrate our approach on a large scale dataset stemming from an online system. 1 Introduction A new family of so-called “Web 2.0” applications is currently emerging on the Web. These include user-centric publishing and knowledge management platforms like Wikis, Blogs, and social resource sharing systems. In this paper, we focus on resource sharing systems, which all use the same kind of lightweight knowledge representation, called folksonomy. The word ‘folksonomy’ is a blend of the words ‘taxonomy’ and ‘folk’, and stands for conceptual structures created by the people. Resource sharing systems, such as Flickr1 or del.icio.us,2 have acquired large numbers of users (from discussions on the del.icio.us mailing list, one can ap￾proximate the number of users on del.icio.us to be more than one hundred thousand) within less than two years. The reason for their immediate success is the fact that no specific skills are needed for participating, and that these tools yield immediate benefit for each individual user (e.g. organizing ones bookmarks in a browser-independent, persistent fashion) without too much overhead. Large numbers of users have created huge amounts of information within a very short period of time. As these systems grow larger, however, the users feel the need for more structure for better organizing their resources. For instance, approaches for tagging tags, or for bundling them, are currently dis￾cussed on the corresponding news groups. Currently, however, there is a lack of theoretical foundations adapted to the new opportunities which has to be overcome. A first step towards more structure within such systems is to discover knowl￾edge that is already implicitly present by the way different users assign tags to resources. This knowledge may be used for recommending both a hierarchy 1 http://www.flickr.com/ 2 http://del.icio.us