Wikipedia reputation talk next week

[Luca de Alfaro]

A Content-Driven Reputation System for the Wikipedia.
Luca de Alfaro
Tuesday, April 24, 2:15pm
Engineering 2 Bldg, Rm 399

We present a content-driven reputation system for Wikipedia authors.
In our system, authors gain reputation when the edits they perform to
Wikipedia articles are preserved by subsequent authors, and they lose
reputation when their edits are rolled back or undone in short order.
Thus, author reputation is computed solely on the basis of content
evolution; user-to-user comments or ratings are not used. The author
reputation we compute could be used to flag new contributions from low-
reputation authors, or it could be used to allow only authors with
high reputation to contribute to controversial or critical pages. A
reputation system for the Wikipedia could also provide an incentive
for high-quality contributions.

We have implemented the proposed system, and we have used it to
analyze the entire Italian and French Wikipedias, consisting of a
total of 691,551 pages and 5,587,523 revisions. Our results show that
our notion of reputation has good predictive value: changes performed
by low-reputation authors have a significantly larger than average
probability of having poor quality, as judged by human observers, and
of being later undone, as measured by our algorithms.

Joint work with Bo Adler.