Right now, no text analytics software exists that can accurately detect sarcasm. Sarcasm is difficult for computers to understand since words are used to convey the opposite of their meaning.
In spoken sarcasm, speakers use tone of voice and body language, which aren't relevant in text. Listeners or readers may also know a speaker's actual opinions. For instance, if Pat knows that Sandy hates spinach, and Pat gets a text message from Sandy saying “My mother made spinach for dinner. YAY! My favorite!”, then Pat is pretty certain that Sandy is being sarcastic. But a computer, like any other stranger, is missing context.
Generally most text data is sincere, so sarcasm is a statistically insignificant problem. However, if you have a data set containing an unusually high level of sarcasm, you may want to use another data set.