Technology Review (07/31/09) Sauser, Brittany
An algorithm capable of ranking the expertise of online users and identifying those who are using a site to spam has been developed by European researchers. The Spamming-resistant Expertise Analysis and Ranking (SPEAR) algorithm assesses users according to a new set of criteria that makes intuitive assumptions about experts. “It distinguishes between ‘discoverers’ and ‘followers,’ focusing on users who are the first to tag something that subsequently becomes popular,” says Cornell University professor Jon Kleinberg. He says SPEAR is a “mutual reinforcement” technique that evaluates popular users and popular content and deems expert users to be the ones who identify the most important content. Users are normally ranked by how often or how recently they add content to the system, and this approach makes the system very susceptible to Web spammers, says Ciro Cattuto at the Complex Network and Systems Group of Italy’s Institute for Scientific Interchange Foundation. He says spammers see the most popular tags and begin loading advertising content with those tags, and SPEAR “performs better than anything currently available–spammers rank very low, their content is not exposed, and eventually they stop polluting the system.” SPEAR uses temporal information to assess user expertise, following the assumption that the people who first discover content that subsequently receives heavy tagging can be classified as trend setters in a community. Conversely, followers find useful information later and tag it because it has already gained popularity.
Filed under: ICT