Sarcasm Detection Methods in Deep Learning: Literature Review

Abstract

Sarcasm is rapidly becoming prevalent in all forms of communication today. Each person has a different way of exploiting and understanding sarcasm. It is difficult for even humans to interpret sarcastic texts, as it depends on a lot of things like perspective, context and tone. This makes it a challenging task to train a machine to distinguish sarcastic text from non-sarcastic text. As there are no concrete rules upon which a model to detect sarcasm can be built, we must resort to promising and upcoming techniques to do so. In this paper, we have reviewed the works done in sarcasm detection using deep learning in combination with the natural language processing techniques. We have also proposed an outline of our own system based on current gaps in literature and challenges in the field.