Twittratr is a StartupWeekend project that started simply with the question of whether tweets about Obama were generally positive or negative in nature. We figured, if we're tracking opinion surrounding Obama, couldn't we do that with any topic? A few of us liked the idea, so we set out to build it.
We wanted to keep things as simple as possible. We built a list of positive keywords and a list of negative keywords. We search Twitter for a keyword and the results we get back are crossreferenced against our adjective lists, then displayed accordingly. There are obvious issues with this, so if you have any ideas on how we could do this better let us know.
We would love to be get access to a semantic analysis tool like Crawdad Technologies, or Visible Technologies. We've sent some emails, but if anyone out there wants to help feel free shoot us an email!