Spear: Spamming‐resistant Expertise Analysis and Ranking in Collaborative Tagging Systems

Abstract

In this paper we discuss the notions of experts and expertise in resource discovery in the context of collaborative tagging systems. We propose that the level of expertise of a user with respect to a particular topic is mainly determined by two factors. Firstly, an expert should possess a high quality collection of resources, while the quality of a Web resource in turn depends on the expertise of the users who have assigned tags to it, forming a mutual reinforcement relationship. Secondly, an expert should be one who tends to identify interesting or useful resources before other users discover them, thus bringing these resources to the attention of the community of users. We propose a graph-based algorithm, SPEAR (SPamming-resistant Expertise Analysis and Ranking), which implements the above ideas for ranking users in a folksonomy. Our experiments show that our assumptions on expertise in resource discovery, and SPEAR as an implementation of these ideas, allow us to promote experts and demote spammers at the same time, with performance much better than the original HITS algorithm and simple statistical measures currently used in most collaborative tagging systems.

"Supertagger" behavior in building folksonomies

“…One noteworthy approach to expert detection is SpammingResistant Expertise Analysis and Ranking (SPEAR) [22,23], a variant of the HITS Web page ranking algorithm [7], that identifies experts according to two principles: First, there should be mutual reinforcement between user expertise and the quality of the annotated items. In other words, an expert user is not only more adept at identifying high quality items, but is also defined by the quality of the items annotated.…”

“…In other words, do prolific taggers demonstrate greater or lesser expertise than non-prolific taggers when annotating items? Detecting expert users in a folksonomy is motivated by an increasing need to distinguish users providing informative contributions from those producing unhelpful contributions (especially spammers) in large folksonomies [6,22].…”

“…Thus, if user m was the first to tag item n, Am,n would be set to the total number of users who tag resource n; but if user m was the last one, then Am,n would be set to 1. Following recommendations by [22,23], the value of Am,n was adjusted by the square root function, such that Am,n = Am,n. Then, user expertise scores per tag are derived by E = Q × A T .…”

Content-Based Recommendation Model in Micro-blogs Community

“…The majority of contributions come from the minority of users. These outstanding users are named as "authority users" [2]. Obviously the influence of authority users is very domain specific; we note that user preference is also domain specific.…”

“…They proposed that the level of expertise of a user in a particular domain is determined by two factors: they possess a high-quality collection of resources and form a mutual reinforcement relationship; they tend to identify interesting or useful resources before other users, thus bringing these resources to the attention of the community of users [2].…”