Sarcasm Detection on Social Networks using Machine Learning Algorithms: A Systematic Review

The researchers have been working rigorously on sarcasm detection on social media over the last decade. Identifying sarcasm in text is very difficult for humans so researchers planned automatic sarcasm detection with computational models. Nowadays people widely use social media to express their views, opinions and comments on others post. The trend of present day is that every person directly or indirectly uses sarcastic words in their day-to-day conversation. Sarcasm is giving a negative meaning with positive words in the conversation. In other words, it is the use of irony to mock or convey contempt or opposite of what you really mean to say. People criticize with positive words. This paper has reviewed different approaches in detection of sarcasm on online social media. An analysis has been performed on each approach as well as noticed drawbacks in it and represented the performance of each approach in terms of precision, recall and accuracy.