Giving wrong result when using negative word in positive way about sentiment HOT 4 CLOSED

I don't know about any other systems that can guess sentiment that are more accurate than AFINN. I think AFINN is generally good enough, but it depends on what you want.
Twitter uses AFINN for trend analysis for example. They know it's sometimes wrong, however even real humans will sometimes mistake the sentiment of a tweet. For Twitter it's more important that it can analyse a lot of tweets quickly, than that it's 100% accurate.
So if you're going to do a lot of analysis, and average the results, the occasional error is not important. If you need high accuracy for a single text, you'll need something else, but I'm not sure wether it exists already, or you'll have to do the research yourself.
Remember, even if you've got a perfect program, in some cases it is impossible to tell if a message was meant sarcastically or as a joke. The Python implementation of AFINN does contain additional support for emoticons.

from sentiment.

Thanks for your help with this @Ghostbird

--

@bansalvks If you are interested in Sentiment Analysis algorithms that properly account for part of speech (POS) I recommend you take a look at some of the foundational research on the subject. AFINN (like nearly all algorithms which form generalizations about natural phenomena) is a heuristic and does not account for negation and many other nuances of human language. Here are a few to get you started:

Opinion Mining and Sentiment Analysis - http://dl.acm.org/citation.cfm?id=1454712
Sentiment analysis of Twitter Data - http://dl.acm.org/citation.cfm?id=2021114

from sentiment.

Not a bug, the result you receive is a perfectly valid AFINN-165 score. Did you look at the paper that this tool is based on?
You expect the software to actually understand human sentiment, which is not what AFINN does. AFINN is a way to guess the sentiment very quickly, in some cases it just guesses wrong.
For your example you will need a much more complex algorithm, which will most likely be slower but make fewer mistakes.

from sentiment.

@Ghostbird Many thanks for replying. I understood what you said and I agree that you are completely right. Could you please guide me in the direction where I could find a solution for my problem, which is to guess the sentiment of the statement (+,-)

from sentiment.