The Pattern library has sentiment analysis built-in, pretty fun toolkit to play around with.<p><a href="http://www.clips.ua.ac.be/pages/pattern-en#sentiment" rel="nofollow">http://www.clips.ua.ac.be/pages/pattern-en#sentiment</a>
Great write-up. My company (Tawlk) actually open sourced a library to automate this very thing. We typically get around 80% accuracy with about 2 million samples.<p>You can grab our sample set here: <a href="https://github.com/downloads/Tawlk/synt/sample_data.bz2" rel="nofollow">https://github.com/downloads/Tawlk/synt/sample_data.bz2</a><p>And check out the project here: <a href="http://github.com/Tawlk/synt" rel="nofollow">http://github.com/Tawlk/synt</a><p>It also ships with a full CLI interface if you just want to play with it without getting knee deep into the code.<p>Also if you want to to see a stripped down stand-alone code sample that steps you through the process I made this gist:<p><a href="https://gist.github.com/1266556" rel="nofollow">https://gist.github.com/1266556</a><p>Enjoy :)
A better example is shown by Jacob Perkins on his blog -
<a href="http://streamhacker.com/2010/05/10/text-classification-sentiment-analysis-naive-bayes-classifier/" rel="nofollow">http://streamhacker.com/2010/05/10/text-classification-senti...</a>