Closed tommedema closed 10 years ago
Hey,
Unfortunately neither of those cases are supported. The idea is to get "good enough" results in enough cases that it's helpful. If you're looking for something that handles context and negation (or just something that gives more accurate results overall), you'll probably need to train your own sentiment classifier over your own dataset.
On Sat, Nov 23, 2013 at 3:29 AM, tommedema notifications@github.com wrote:
Can your algorithm deal with negation? E.g.
"I am not happy" should be negative rather than positive.
Also, can it deal with sentiment analysis wrt a specific context? E.g.
"Horses are better than dolphins." is positive in the context of horses but negative in the context of dolphins.
— Reply to this email directly or view it on GitHubhttps://github.com/thinkroth/Sentimental/issues/8 .
Hmm, that's a shame.
Do you happen to know of algorithmic software that is more sophisticated and that I could run on my Linux server? E.g. Java based with command line arguments or an API.
Thanks.
Can your algorithm deal with negation? E.g.
"I am not happy" should be negative rather than positive.
Also, can it deal with sentiment analysis wrt a specific context? E.g.
"Horses are better than dolphins." is positive in the context of horses but negative in the context of dolphins.
Finally, can it deal with expressions? E.g.
"The NYSE hit an all time high today." is very positive