Sarcasm is the use of language where people used positive words in order to convey a negative message. Sarcasm is an
example of what some researchers call uncertain speaking, where what is said by the people is actually differ from what is meant
(especially so as to insult or wit somebody, to indicate irritation, or to be funny). sentiment analysis on social media has been
extensively used, however it's still rare to look out sentiments and analyse them, considering the detection of satire in it. Satire
detection in sentiment analysis is a difficult task to do. The most used way for sarcasm detection is Machine learning approach.
This paper proposed the used of linear and radial basis function kernels of support vector machine algorithm for sarcasm
detection and compared them to find out which kernel provide best accuracy.
