Also in this playlist...
This transcript is automatically generated
Have brought you many debt we say precocious young people on this program and today we -- joined by another -- He created a program that tracks what people -- saying about the presidential candidates on Twitter.
And he turns it into a -- so this could be the future appalling maybe eight joining us now is.
If someone writes I really love Obama all I'm voting for Romney and divided on -- your.
Gives those mentions economists call and -- -- -- a -- I'm I'm I'm I'm trying to get to grips with what you do and I'm right -- is this -- so it takes it takes each individual -- Mentioning either candidate and -- scores on a scale of one to negative one.
So it's doing that 101000 times and -- quarter million times a day so the whole group for the actual -- them coming up with is exponentially bigger than.
Normal rules -- your checking every tweet that goes out and you can use for key -- -- and expressions.
And then you if you bring it altogether ten dollars and -- times and -- -- you got an algorithm that does that.
And creates a statistic which is a form of a poll -- Okay that's what you do basically yes so -- the way it works is that it.
I rode up the sentiment.
Analysis -- promotions I did before actually -- application and I said oh you know -- And reduce the candidates -- plugged in and I let it run and run on the computer and -- -- national -- That what you wanna do that pat you could theoretically sell it kind of my -- I'm actually right now working on.
Implementing in different ways so.
With movies Blockbuster had seconds to projecting where that's that's gonna go based on sense of public since Osce could do a poll on a new movie people like I don't know lower before -- -- is the buzz is a positive and negative going after the release that movie and it would be instant and it -- you flat -- poll.
I mean literally by the second yet there's no delay if you look in the -- for the actual poll data.
It's there's a thirty minute value we value among -- -- you have you have different sets of data.
Thirty minutes and us media and that's recent months fluctuations and can you give me an idea of of any kind of polling advantage of bomb of those is roaming what what -- tweet to polls saying now as of right now it's -- Romney -- roughly 2% -- That has been changing over the past month of this time last month Obama had 33%.
With which matches what the national debt polls are showing a slight movement.
-- -- -- -- -- -- -- Mean which is which I was I had I tested throughout the Republican primaries and that's I saw the results then and now under food but how much would you want it to to -- thankful I and you -- moving more come on -- you -- but this is a financial program initial.
Couple million couple million settled for as low as there.
Little -- -- laughing well it's I've got a summer job appearing to follow politics that's right he's -- again.
It's your math genius it's something that's always come naturally to me and I would -- around with that one -- saw the Obama Trent saw the volume of Obama to shining and -- and thought why Canada.
And -- positive or negative and can you know -- possesses you're on -- something quite clearly and we thank you very much for being with us today -- -- good -- young man -- find stuff appreciate it thank.
Filter by section