Dear John and all,

With regard to your discussion on examining Wikipedia article quality, the idea somehow overlaps with our intuition of measuring article quality based on the authority of authors/reviewers. We have been researching from this perspective. Below attached is the abstract of our research paper, titled "Measuring Article Quality in Wikipedia: Models and Evaluation", published in CIKM 2007:

" Wikipedia has grown to be the world largest and busiest free encyclopedia, in which articles are collaboratively written and maintained by volunteers online. Despite its success as a means of knowledge sharing and collaboration, the public has never stopped criticizing the quality of Wikipedia articles edited by non-experts and inexperienced contributors. In this paper, we investigate the problem of assessing the quality of articles in collaborative authoring of Wikipedia. We propose three article quality measurement models that make use of the interaction data between articles and their contributors derived from the article edit history. Our Basic model is designed based on the mutual dependency between article quality and their author authority. The PeerReview model introduces the review behavior into measuring article quality. Finally, our ProbReview models extend PeerReview with partial reviewership of contributors as they edit various portions of the articles. We conduct experiments on a set of well-labeled Wikipedia articles to evaluate the effectiveness of our quality measurement models in resembling human judgment. "

We would appreciate your comments and suggestion.

Regards, Meiqun HU

-----Original Message-----

Date: Mon, 10 Mar 2008 23:00:39 +0100 From: John Erling Blad john.erling.blad@jeb.no Subject: [Wikiquality-l] Checking article quality through number of visitors To: wikiquality-l@lists.wikimedia.org Message-ID: 47D5AF87.2030501@jeb.no Content-Type: text/plain; charset="iso-8859-1"

Is there anyone that has done any research on how the number of visitors relates to the article quality? I believe it is related somehow but I'm not sure how it can be modeled. It works by counting the visitors that reads a particular segment of the article, and then will accept the particular segment as correct when a sufficient number of visitors has been visiting. It can work together with a system for writer grading, were this system will change the grade from whatever the writer has.

Compared to this a "stable versions" is like having a visitor with ultimate power to mark the revision as good. This system does not give the visitors such ultimate power, and in fact will not give give them more than a small fraction of the power necessary to claim the revision is free of vandalism. Combined I guess it is possible to make a system that will be better than anyone of them alone.

Any real vandalism will most likely never be marked as good, because the limit can be set so high that it will be found by someone long before it is marked as "patrolled", and then most likely nothing or very little of the revision will survive so the revision itself will never be marked as patrolled. If a known good writer contributes a revision, then it will get a flying start and it will need few visitors ("anonymous patrollers") before it is marked as "good". If the writer is unknown the revision will need a lot of visitors before it is marked as good.

Even very seldom read articles have several visitors each week, and through a year this will add up to a considerable amount of visitors.

John