Ontology Learning from Folksonomies
Organizers:
Andreas Hotho
Data Mining and Information Retrieval Group
University of Würzburg
hotho@informatik.uni-wuerzburg.de
Robert Jäschke (main contact)
Knowledge and Data Engineering Group
University of Kassel
jaeschke@cs.uni-kassel.de
Ontologies and the Web 2.0 are two fields of research that allow for rich and intriguing interactions. The most important characteristic of the Web 2.0 is the diminishing distinction between content provider and content consumer. This enables millions of users to collaboratively publish content online and often annotate it with freely chosen keywords so called `tags'. Ontology learning from folksonomies allows us to make the knowledge represented by folksonomies more explicit and to foster the `wisdom of the crowds' for the creation of ontologies.
This half day tutorial targets a broad audience in the knowledge engineering community. The goal of the tutorial is to acquaint the reader with the principal problems and solutions to learn ontologies from folksonomies. The lecture gives an overview on ontology learning methods for the Web 2.0 with an emphasis on folksonomies and collaborative tagging systems. The second part of the tutorial focusses on ontology learning approaches. The application of association rules on different projections of the hypergraph allows to extract relations on all modes of a folksonomy, namely tags, users, and resources, with the goal of extracting more semantics and gaining insights into the behavior of the users.