We're Being Studied!

How's that for role reversal on Science Blogs?

Well, leave it to the good folks at Carnegie Mellon...

Scientists have long studied how information, influence or physical items move through networks. But by combining that field of research with how to optimally detect the flow in a cost-effective way, the Carnegie Mellon researchers have devised a formula, or algorithm, that could lead to dramatically improved sensor networks, whether geared toward political blogs or posture.

But how would this cascade be modeled? What sensors, or blogs in this case, should be tapped to maximize the likelihood of capturing a big story early on during its propagation over the blogosphere?

Researchers came up with a complex mathematical equation called the cost-effective lazy forward-selection algorithm, later dubbed the Cascades algorithm. It seeks to maximize reward (for blogs - detecting the most news in the least amount of time) and minimize cost (the time spent reading blogs). By utilizing the law of diminishing returns, the outcomes were calculated lists of top blogs (different results depended on the number of blogs budgeted in).

Who made the lists? Scienceblogs sure did...
Top 100
Top 5,000

Neat, eh? If the results can be trusted, imagine the possibilities for this kind of information applied both online and in other complex systems well beyond the blogosphere...

i-2dc6b61cd84c380335809490774ef0bd-blogosphere.jpg

The blogosphere as social network

More like this

"If the results can be trusted.."

Yes, indeed. There's the hitch. Detecting the "the most news in the least amount of time" is a far cry from detecting the most trustworthy news. As we saw with the pre-war reporting on Iraqi WMD [sic], the news item that appears most frequently may not be correct. A popularity contest need have nothing to do with underlying reality.

But for those who would trust a computer algorithm to decide the latter -- ie, which information on the internet is most trustworthy (something the Carnegie Mellon algorithm does not do) -- I do have some prime sea coast property in Arizona just waiting to be enjoyed.

By Dark Tent (not verified) on 29 Jan 2008 #permalink

Oh no, not the dreaded "web bots". The are the techno-nerd answer to the bible code and about as worthwhile.

Most people I know on LiveJournal are quite active in other parts of the Internet, but they don't tend to link between their Web sites and their LJ. LiveJournal is, for many of us, meant to be its own little island.

Scienceblogs is #98. Michelle Malkin is #5. What is wrong with this picture? I'm not sure how much accurate news you could get from her.

I have to wonder about their definition of blog. Guess who's listed at #271 on the longer list? Lots of news, but hardly what I'd call a blog.