News for hackers. Stuff that matters.
Re:Stanford researchers to open source model they say has na... by aantix on Friday October 04, 2013 @12:09AM (#6493670)
I feel like Borat for doing this, but I entered :

"I loved this movie.. NOT!"

and it classified it as positive. :)

Re:Stanford researchers to open source model they say has na... by koyote on Friday October 04, 2013 @09:25AM (#6495174)
I did the same!

It also doesn't seem to like swear words and web abbreviations (like j/k for example). Perhaps with more data (from twitter or similar) it could go a long way though; as it definitely needs to learn about the more uncouth words.

Re:Stanford researchers to open source model they say has na... by vivekn on Friday October 04, 2013 @04:51AM (#6494389)
Thwarted expectations and sarcasm are difficult to handle. You are down to the choice of optimizing for this particular case or the majority of text.
Re:Stanford researchers to open source model they say has na... by gojomo on Friday October 04, 2013 @02:22AM (#6493988)
You've found the 15% it gets wrong. Though, given the other tricky cases it does well on, more training data might help.
Re:Stanford researchers to open source model they say has na... by saraid216 on Friday October 04, 2013 @01:02AM (#6493805)
Borat is not a measure by which any sentiment analysis should be done... :P