Not logged in. · Lost password · Register

All content © NFGworld, unless otherwise noted, except for stuff we stole. Contact the editor-in-chief : baldbutsuave@thissitesdomain, especially if you are an attractive young female willing to do nude photography modelling. All rights reversed. 703

NFG (Administrator) #1
Member since May 2011 · 2485 posts · Location: Brisbane
Group memberships: Administrators, Members
Show profile · Link to this post
Subject: Predictive modelling and flash trends...
It would seem that any algorithm designed to predict trends is in danger of being outpaced by the things it purports to track.  How can it be possible to watch for emerging trends when things like SMS, IM and Twitter make 'flash trends' boom and bust faster than the watchers can even notice, let alone evaluate?  Or is this even a consideration?

That's the crux of a recent techdirt post.

Quote by techdirt:
I'd say I'm not at all convinced (as the article posits) that Twitter "confounds" predictive modeling. I think it still requires other random events to occur -- and those have always happened in the past. Perhaps the interconnected nature of the world today can massively amplify a sudden flash fad, but that doesn't necessarily mean you toss predictive modeling out the window.

Or maybe not:

Quote by techdirt:
and (surprise!) it turns out that the guy who wrote the article runs a company selling a tool that competes with predictive modeling software focusing on "behavioral analytics."

So, much like climate change, this means there will be random spikes of extreme behavior that cannot be modelled, but given the inherently random nature of the future, wasn't it always thus?

Random things will continue to be random, predictable things will continue to be predicted.  Move along, nothing new to see here.
Close Smaller – Larger + Reply to this post:
Smileys: :-) ;-) :-D :-p :blush: :cool: :rolleyes: :huh: :-/ <_< :-( :'( :#: :scared: 8-( :nuts: :-O
Special characters:
We love UNB by Yves Goergen!