Truemors
is reporting from
Member
NP Rank:
NP Rank:
Algorithm Rates Trustworthiness of Wikipedia Pages
Share: 
by AlanEvans | August 31, 2007 at 03:56 am
521 views | 10 Recommendations | 3 comments
Researchers at UCSC developed a tool that measures the trustworthiness of each Wikipedia page. Roughly speaking, the algorithm analyzes the entire 7-year user-editing-history and utilizes the longevity of the content to learn which contributors are the most reliable: If your contribution lasts, you gain 'reputation,' whereas if it's edited out, your reputation falls. The trustworthiness of a newly inserted text is a function of the reputation of all its authors, a heuristic that turned out to be successful in identifying poor content.
This actually just gets more and more cool the more you look at it. Here is a link to the project's demo site:
http://enwiki-trust.cse.ucsc.edu/index.php/Special:Random
What you see is a random page from the test site (of about 100 pages) with a kind of heat-map overlayed showing you the "trust level" of each edit that has been made on the page. Text on a white background is trusted, whereas text on an orange background less so. The shade of background color reflects the level of trust.
Particularly fun is if you click the "history" tab of any page, and compare to historical revisions of the page. The tool then shows you the increases and decreases in trust level of every section of the page as it changes.
Click the "random page" link in the left margin to see how other pages compare in terms of trust level.
It could be very interesting to see a system like this operating on wikipedia itself.
What is NowPublic?
NowPublic lets people work together to cover news events around the world.
Most RecentMost Recommended Comments (3)
at 04:47 on August 31st, 2007
I have always found Wiki to be as accurate as I need it to be. It provides me with a good summary of what I'm looking for and it includes links to external sites.
So long as you know its limitations, I think Wiki is a very valuable resource.
at 05:31 on August 31st, 2007
I find that, the more obscure the information, the less likely I am to trust the Wiki source, as it's less likely to be trafficked by other experts in the field. I guess I'm just paranoid.
at 10:11 on August 31st, 2007
AlanEvans, I like this story. I don't know how effective this algrotihm in finding trustwrothy parts, or how many false alerts it generates, but it has some cool ideas, and the reperesentation also looks cool.