Review of English literature on figurative language applied to social networks

For a long time, figurative language was studied merely from linguistic perspectives, yet it has lately captured the attention of other fields, such as natural language processing, sentiment analysis, and machine learning. The increasing interest in figurative language calls for a clear overview of figurative language research. To address this need, we present a review of English literature on figurative language applied to social networks in a five-year period: from 2013 to 2017. The aim of this review is to identify the most commonly researched figurative devices, as well as their discriminant features, detection approaches and methods, and languages in which they are studied. To this end, we analyze and evaluate 521 research works and present 45 primary studies. The results show that sarcasm is the most studied figurative device, with 56% of the total frequencies. Also, 87% of the studies are based on the supervised machine learning approach, and the support vector machine classifier has been the most used to detect the different types of figurative language (i.e., figurative devices). Similarly, more than half of the literature focuses on figurative language in English.