Real Time Sarcasm Detection on Twitter using Ensemble Methods

Abstract

Sarcasm means saying the opposite of what you mean in order to make fun of someone and a type of humour that responds to a situation. Sarcasm reorganisation approach is quite beneficial to enhancing automated sentiment analysis data from microblogging and social media sites. The term “sentiment analysis” relates to the study of internet users reported feelings and viewpoints in a particular group, as well as their identification and aggregation. One of the most complicated problems in sentiment analysis is detecting sarcasm. It's a tough task to classify sarcastic sentence forms. This work uses two hybrid machine learning approaches, namely Stacked Generalization and Boosting ensemble methods with Support Vector Machine (SVM), Random Forest (RF) and KNN as base classifiers and Logistic Regression (LR) as Meta classifiers to detect real-time sarcasm on Twitter.