Social bookmarking systems and their emergent information structures, known as folksonomies, are increasingly important data sources for Semantic Web applications. A key question for harvesting semantics from these systems is how to extend and adapt traditional notions of similarity to folksonomies, and which measures are best suited for applications such as navigation support, semantic search, and ontology learning. Here we build an evaluation framework to compare various general folksonomy-based similarity measures derived from established information-theoretic, statistical, and practical measures. Our framework deals generally and symmetrically with users, tags, and resources. For evaluation purposes we focus on similarity among tags and resources, considering different ways to aggregate annotations across users. After comparing how tag similarity measures predict user-created tag relations, we provide an external grounding by user-validated semantic proxies based on WordNet and the Open Directory. We also investigate the issue of scalability. We find that mutual information with distributional micro-aggregation across users yields the highest accuracy, but is not scalable; per-user projection with collaborative aggregation provides the best scalable approach via incremental computations. The results are consistent across resource and tag similarity.
C. Scholz, J. Illig, M. Atzmueller, und G. Stumme. 25th ACM Conference on Hypertext and Social Media (to appear), Santiago, Chile, September 1-4, ACM, (2014)
M. Atzmueller, A. Ernst, F. Krebs, C. Scholz, und G. Stumme. 5th International Workshop on Modeling Social Media: Mining Big Data in Social Media at the 23rd International World Wide Web Conference, WWW 2014, Seoul, South Korea, (2014)
C. Scholz, M. Atzmueller, und G. Stumme. 5th International Workshop on Modeling Social Media: Mining Big Data in Social Media at the 23rd International World Wide Web Conference, WWW 2014, Seoul, South Korea, (2014)
M. Becker, J. Mueller, A. Hotho, und G. Stumme. 1st International Workshop on Pervasive Urban Crowdsensing Architecture and Applications, PUCAA 2013, Zurich, Switzerland -- September 9, 2013. Proceedings, Seite 1175--1182. New York, NY, USA, ACM, (2013)
J. Mueller, S. Doerfel, M. Becker, A. Hotho, und G. Stumme. Recommender Systems and the Social Web Workshop at 7th ACM Conference on Recommender Systems, RecSys 2013, Hong Kong, China -- October 12-16, 2013. Proceedings, Seite New York, NY, USA. ACM, (2013)accepted for publication.
F. Mitzlaff, M. Atzmueller, G. Stumme, und A. Hotho. Complex Networks IV, Volume 476 von Studies in Computational Intelligence, Springer Verlag, Heidelberg, Germany, (2013)
M. Kibanov, M. Atzmueller, C. Scholz, und G. Stumme. Cyber, Physical and Social Computing (CPSCom), 2013 IEEE International Conference on, Los Alamitos, CA, USA, IEEE Computer Society, (2013)
C. Scholz, M. Atzmueller, M. Kibanov, und G. Stumme. Advances in Social Networks Analysis and Mining (ASONAM), 2013 International Conference on, Los Alamitos, CA, USA, (2013)
F. Mitzlaff, und G. Stumme. (2013)cite arxiv:1303.0484Comment: Historically, this is the first paper on the analysis of names in the context of the name search engine 'nameling'. arXiv admin note: text overlap with arXiv:1302.4412.
J. Mueller, S. Doerfel, M. Becker, A. Hotho, und G. Stumme. Recommender Systems and the Social Web Workshop at 7th ACM Conference on Recommender Systems, RecSys 2013, Hong Kong, China -- October 12-16, 2013. Proceedings, 1066, Aachen, Germany, CEUR-WS, (2013)
C. Scholz, M. Atzmueller, und G. Stumme. Proceedings of the 2012 ASE/IEEE International Conference on Social Computing and 2012 ASE/IEEE International Conference on Privacy, Security, Risk and Trust, Seite 312--321. Washington, DC, USA, IEEE Computer Society, (2012)