RSDC 08
Using Co-occurence of Tags and Resources to identify Spammers
Abstract
Today, more and more social networking websites support
collaborative tagging, which allows users to annotate
resources (e.g., video clips, blog posts, and bookmarks) on the
web. Due to its increasing popularity, however, spammers started
to target this new type of service and generate misleading tags
either to increase the visibility of some resources or simply to
confuse users. Consequently, the performance of applications built
upon tag data, such as recovery and discovery of web resources,
can be limited. In this paper, we propose an algorithm to identify
spammers from the collaborating systems by employing a
spam score propagating technique. The three dimensional
relationship among users, tags and web resources is firstly
represented by a graph structure. A set of seed nodes, where each
node represents a user, are then selected and assigned values to
indicate whether the corresponding users are spammers or not. The
initial values are propagated through the graph to infer the
status of the remaining users. Our experimental results
demonstrate the effectiveness of this approach in identify tag
spammers.
Full Paper
RSDC08.pdf
BibTex Entry
