Connectivity on Complete Lattices
Journal of Mathematical Imaging and Vision
An open graph visualization system and its applications to software engineering
Software—Practice & Experience - Special issue on discrete algorithm engineering
Usage patterns of collaborative tagging systems
Journal of Information Science
InfoScale '06 Proceedings of the 1st international conference on Scalable information systems
Combating spam in tagging systems
AIRWeb '07 Proceedings of the 3rd international workshop on Adversarial information retrieval on the web
Fighting Spam on Social Web Sites: A Survey of Approaches and Future Challenges
IEEE Internet Computing
Network properties of folksonomies
AI Communications - Network Analysis in Natural Sciences and Engineering
Weighted graphs and disconnected components: patterns and a generator
Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining
The anti-social tagger: detecting spam in social bookmarking systems
AIRWeb '08 Proceedings of the 4th international workshop on Adversarial information retrieval on the web
Hi-index | 0.00 |
Data created by social bookmarking systems can be described as 3-partite 3-uniform hypergraphs connecting documents, users, and tags (tagging networks), such that the toolbox of complex network analysis can be applied to examine their properties. One of the most basic tools, the analysis of connected components, however cannot be applied meaningfully: Tagging networks tend to be almost entirely connected. We therefore propose a generalization of connected components, m-hyperincident connected components. We show that decomposing tagging networks into 2-hyperincident connected components yields a characteristic component distribution with a salient giant component that can be found across various datasets. This pattern changes if the underlying formation process changes, for example, if the hypergraph is constructed from search logs, or if the tagging data is contaminated by spam: It turns out that the second- to 129th largest components of the spam-labeled Bibsonomy dataset are inhabited exclusively by spam users. Based on these findings, we propose and unsupervised method for spam detection.