Explicit sarcasm handling in emotion level computation of tweets - a big data approach

Abstract

Social media like Twitter offers an important window into the emotions of those who use the platform to share opinions on various topics. Nearly 79% of the world population use social media to express their opinions on various topics. Various commercial organizations like E-commerce sites, health departments, disaster management activities, etc. may want to compute the emotion levels of tweets for analyzing and gaining useful insights into the user's opinions and preferences and using the result of the analysis for various purposes like determining social influence, information diffusion modeling, sentiment analysis, etc. The existing tools for computing the emotion level polarity, however, do not consider sarcasm that most predominantly exist in short texts like tweets. This paper presents a big data approach for computing emotion levels of each tweet for a given day, with handling of explicit sarcasm in tweets. The goal is to provide an efficient and, at the same time, a scalable approach for computing emotion levels in tweets.