Google even searches the future now

Wired already reported that Google Trends could have been used to find out about the Swine Flu epidemic in Mexico weeks before it was reported in the news media. Then, in anticipation of the Eurovision Song Contest 2009, the google engineers created a widget that would take Google Trends data as input (per country), and transform the search activity in each country to Eurovision points of 1 to 12. I copied the prediction to my Facebook page just when the Eurovision final was starting:

1 Norway (Alexander Rybak) 388
2 Turkey (Hadise) 358
3 Greece (Sakis Rouvas) 277
4 Ukraine (Svetlana Loboda) 197
5 Sweden (Malena Ernman) 188
6 France (Patricia Kaas) 165
7 Russia (Anastasia Prikhodko) 126
8 United Kingdom (Jade Ewen) 74
9 Denmark (Brinck) 43
10 Switzerland (Lovebugs) 40

And the real results:

1 Norway 387
2 Iceland 218
3 Azerbaijan 207
4 Turkey 177
5 United Kingdom 173
6 Estonia 129
7 Greece 120
8 France 107
9 Bosnia & Herzegovina 106
10 Armenia 92

Ok, so let's be clear about this: Google was a bit lucky here. Only 50% of the points this year were awarded by public televoting. The other half was awarded by the return of the infamous national committees, which really could vote for anything. But still, strong performance from Google, since Turkey was 4th and Greece 7th. Iceland and Azerbaijan otoh completely flew under Google's radar!

If I had commented on this before the contest, I would have said that Google's results are skewed to favor the top hits and should be somehow scaled. This is because until Saturday, the record amount of points received by an ESC winner was Lordi with 292, so to predict someone getting close to 400 points seemed ridiculous, until Norway got it! Behind Norway though the others got less, which is explained by the rise of those countries that didn't show up on Google's list.

And how is this interesting? Data-mining is the key word. There's lots of data out there that can predict lots of things, if you just bothered to look into it.

We used to say that stock market and/or gambler's provide good data for prediction. This was also true this time with Norway and Turkey having among the lowest odds too - and again Iceland and Azerbaijan suspiciously missing. So it seems Google Trends and gamblers produced fairly similar predictions being right and wrong on same accounts.

And finally, the tabloid press was up to its standards with the Finnish headlines having "Finland rising to favorite in Eurovision". How did we end up among the 25 contestants:

25 Finland 22

...but even I could have predicted that!

Add new comment

The content of this field is kept private and will not be shown publicly.
  • No HTML tags allowed.
  • External and mailto links in content links have an icon.
  • Lines and paragraphs break automatically.
  • Web page addresses and email addresses turn into links automatically.
  • Use [fn]...[/fn] (or <fn>...</fn>) to insert automatically numbered footnotes.
  • Each email address will be obfuscated in a human readable fashion or, if JavaScript is enabled, replaced with a spam resistent clickable link. Email addresses will get the default web form unless specified. If replacement text (a persons name) is required a webform is also required. Separate each part with the "|" pipe symbol. Replace spaces in names with "_".
About the bookAbout this siteAcademicAmazonBeginnersBooksBuildBotBusiness modelsbzrCassandraCloudcloud computingclsCommunitycommunityleadershipsummitConsistencycoodiaryCopyrightCreative CommonscssDatabasesdataminingDatastaxDevOpsDrizzleDrupalEconomyelectronEthicsEurovisionFacebookFrosconFunnyGaleraGISgithubGnomeGovernanceHandlerSocketHigh AvailabilityimpressionistimpressjsInkscapeInternetJavaScriptjsonKDEKubuntuLicensingLinuxMaidanMaker cultureMariaDBmarkdownMEAN stackMepSQLMicrosoftMobileMongoDBMontyProgramMusicMySQLMySQL ClusterNerdsNodeNoSQLodbaOpen ContentOpen SourceOpenSQLCampOracleOSConPAMPPatentsPerconaperformancePersonalPhilosophyPHPPiratesPlanetDrupalPoliticsPostgreSQLPresalespresentationsPress releasesProgrammingRed HatReplicationSeveralninesSillySkySQLSolonSunSybaseSymbiansysbenchtalksTechnicalTechnologyThe making ofTungstenTwitterUbuntuvolcanoWeb2.0WikipediaWork from HomexmlYouTube

Search

Recent blog posts

Recent comments