For those of you interested in reputation management systems, I wrote an overview for Dr. Hersh’s Information Retrieval course as part of my Biomedical Informatics Graduate Certificate. I think it turned out well- here’s the abstract:
As more people turn to the Internet for information, they may find themselves reading material from, or participating in, online communities. It is more important than ever that one is able judge the quality of the members of these communities. The paper looks at nineteen online communities, which offered reputation management systems. By including both healthcare and non-healthcare related sites, a variety of processes were noted. There are both simple and quite sophisticated systems. Domain experts were not only directly identified by use of external credentials, but also based on a synthesis of the member’s thoughts as to the individual’s contribution to the information within the community. No system appeared to be the best, but many that seemed to fit their particular community needs. Reputation management included self-reporting, external credentials, voting, surveys, metamoderation, levels, additional powers within the community, means of motivation, and chance. Along with the overview, there is a discussion of issues regarding simple Information Retrieval approaches as well as weighting, granularity, consistency, evaluation, and when things go wrong. Evaluation of these systems was not part of this overview.