I can’t form a team in this Hackathon, due to time limitations. But I like to share this idea and if a team likes to pick up I’d be glad to support you when it comes to data architecture / schema design.
This is more a fun project but it might show interesting aspects of data mining.
The question is whether certain colors occur more frequently in certain languages / regions / cultures and whether a reliable correlation can be proven.
E.g. (completely personal opinion) I get the impression of “green” when I think of Ireland. The sources which point to a certain color might be of interest, too. Are these daily news messages, pictures, text or tourism infos, campaigns?
some analytical skills + various