Tuesday, 23 November 2010

Can the Xfactor result be predicted from social network activity?

Or specifically can facebook be used a tool for prediction and if so, is it effective?
Click the image to see it in full, sorry it takes up so much space!
I've used a similar technique to that I used for the elections earlier this year (see link on the right), rushing round facebook pages and noting down the amount of users that are fans of each of the artists. I thought I'd start off highlighting all those who made it to the final 16(who are all abreviated to their first four letters). On the left are those still in the competition, and on the right those who have been eliminated. Note I have taken a log scale of the number of fans so that comparisons can be made, as the range of data was large (cher had 1,101,203; whereas poor john only had  6325). What do we see?

  • Cher's fanpage is vastly higher than that of any other contestant (and is in fact more than a 1/3 of the size of mentor Cheryl's fanpage). Why?
  • Of the contestants still battling it out Katie has the fewest fans, even lower than some of those evicted, is this why she has been in the sing-off so often?
  • Diva Fever and Aiden have more fans than some of those still in the competition. Did people only start to miss these acts once they'd left? Should they have stayed in for longer?
I wish I'd started collecting data earlier as the systems dynamics could have been analysed and attributed to real life events and would perhaps be able to make some comments or conclusions.
I plan to collect more data on those still in the competition up till it's finished and posting it here, but at a first impression of facebook data it looks like Cher Lloyd has the Xfactor in the bag.


NOTE: It must be noted also that not all people vote and it may be that only a few people(in relation to total fans) are dictating the decisions of the competition, that people may be fan of more than one contestant and that people can vote tactically - by not voting, voting for a contestant who isn't their favourite, or by voting for several contestants. Age structure probably plays a part in the voting system. Conclusions taken from the figure do not take these factors into account.

Saturday, 13 November 2010

Climate Change: is it? Part 1.

Despite the huge amount of funds, science and political/social movements, there are still so called 'climate deniers' around; probably the most famous of whom is Jeremy Clarkson. There skepticism was pushed under the carpet by many, until last years 'climategate scandal' when University of East Anglia emails were hacked and leaked on the internet. Not so bad, until it was found these emails showed that data used in papers that predict some aspects of climate change and in turn used to push government and organistaional policy and spending were of poor standard, used questionable methods and were stored in an unorganised way. This scandal has got climate scientists quaking in their boots as organisations, mostly US, employ people to meticulously look through climate research and try to unroot and disprove it.

Standards of climate research through data collections, simulations and interepatations has obviously had to increase and many of those working in the area have had to be more careful in what they both say and write to avoid being put in a compromising position or being taken the wrong way.

Climate is defined as the average weather conditions for a region over a reasonable time period. Hardly a robust at all. But it is easy to see that skeptics with logical arguements are not argueing that climate does not change- because invariablly it does;  it doesn't take a human lifetime to see one winter is different to the next.  Rather they are more concerned with the qualitative and quantitive results about future climate predictions - especially with the well known unreliability of weather predictions (though they are usually fairly accurate most of the time!) and whether human life and its activities are a big enough force to change the course of nature itself.

So what makes this area so controversial? Two major reasons I think:

1. Price - To revert anthropogenic climate change as models predict will require a lot of money and policy change and is time, money, effort best spent in this area?
"oh hard times,
for the prize, yes the prize,
i thought i knew
oh it's the price we gotta pay
and all the games we gotta play
makes me wonder if it's worth it to carry on" The Price by Twisted Sister.

"There are no solutions...there are only tradeoffs" Thomas Sowell, Economist.

2. Complexity - Earth Systems Science isn't easy, especially when it's spatio-temporal bounds are large (variations over time and space differ widely).

To put it simply:
"You never really know what you can't really see" Relief by Cold War Kids.
Or the more complex version:
"Reports that say that something hasn't happened are always interesting to me, because as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns -- the ones we don't know we don't know." Donald Rumsfeld, previous US Defense Secretary.